Funding, Deals & Partnerships: BIOLOGICS & MEDICAL DEVICES; BioMed e-Series; Medicine and Life Sciences Scientific Journal – http://PharmaceuticalIntelligence.com
Group of Researchers @ University of California, Riverside, the University of Chicago, the U.S. Department of Energy’s Argonne National Laboratory, and Northwestern University solve COVID-19 Structure and Map Potential Therapeutics
Reporters: Stephen J Williams, PhD and Aviva Lev-Ari, PhD, RN
This illustration, created at the Centers for Disease Control and Prevention (CDC), reveals ultrastructural morphology exhibited by coronaviruses. Note the spikes that adorn the outer surface of the virus, which impart the look of a corona surrounding the virion, when viewed electron microscopically. A novel coronavirus virus was identified as the cause of an outbreak of respiratory illness first detected in Wuhan, China in 2019.
Image of newly mapped coronavirus protein, called Nsp15, which helps the virus replicate.
How UC is responding to the coronavirus (COVID-19)
The University of California is vigilantly monitoring and responding to new information about the coronavirus (COVID-19) outbreak, which has been declared a global health emergency.
The 3-D structure of a potential drug target in a newly mapped protein of COVID-19, or coronavirus, has been solved by a team of researchers from the University of California, Riverside, the University of Chicago, the U.S. Department of Energy’s Argonne National Laboratory, and Northwestern University.
The scientists said their findings suggest drugs previously developed to treat the earlier SARS outbreak could now be developed as effective drugs against COVID-19.
The initial genome analysis and design of constructs for protein synthesis were performed by the bioinformatic group of Adam Godzik, a professor of biomedical sciences at the UC Riverside School of Medicine.
The protein Nsp15 from Severe Acute Respiratory Syndrome Coronavirus 2, or SARS-CoV-2, is 89% identical to the protein from the earlier outbreak of SARS-CoV. SARS-CoV-2 is responsible for the current outbreak of COVID-19. Studies published in 2010 on SARS-CoV revealed inhibition of Nsp15 can slow viral replication.This suggests drugs designed to target Nsp15 could be developed as effective drugs against COVID-19.
Adam Godzik, UC Riverside professor of biomedical sciences Credit: Sanford Burnham Prebys Medical Discovery Institute
“While the SARS-CoV-19 virus is very similar to the SARS virus that caused epidemics in 2003, new structures shed light on the small, but potentially important differences between the two viruses that contribute to the different patterns in the spread and severity of the diseases they cause,” Godzik said.
The structure of Nsp15, which will be released to the scientific community on March 4, was solved by the group of Andrzej Joachimiak, a distinguished fellow at the Argonne National Laboratory, University of Chicago Professor, and Director of the Structural Biology Center at Argonne’s Advanced Photon Source, a Department of Energy Office of Science user facility.
“Nsp15 is conserved among coronaviruses and is essential in their lifecycle and virulence,” Joachimiak said. “Initially, Nsp15 was thought to directly participate in viral replication, but more recently, it was proposed to help the virus replicate possibly by interfering with the host’s immune response.”
Mapping a 3D protein structure of the virus, also called solving the structure, allows scientists to figure out how to interfere in the pathogen’s replication in human cells.
“The Nsp15 protein has been investigated in SARS as a novel target for new drug development, but that never went very far because the SARS epidemic went away, and all new drug development ended,” said Karla Satchell, a professor of microbiology-immunology at Northwestern, who leads the international team of scientists investigating the structure of the SARS CoV-2 virus to understand how to stop it from replicating. “Some inhibitors were identified but never developed into drugs. The inhibitors that were developed for SARS now could be tested against this protein.”
Rapid upsurge and proliferation of SARS-CoV-2 raised questions about how this virus could become so much more transmissible as compared to the SARS and MERS coronaviruses. The scientists are mapping the proteins to address this issue.
Over the past two months, COVID-19 infected more than 80,000 people and caused at least 2,700 deaths. Although currently mainly concentrated in China, the virus is spreading worldwide and has been found in 46 countries. Millions of people are being quarantined, and the epidemic has impacted the world economy. There is no existing drug for this disease, but various treatment options, such as utilizing medicines effective in other viral ailments, are being attempted.
Godzik, Satchell, and Joachimiak — along with the entire center team — will map the structure of some of the 28 proteins in the virus in order to see where drugs can throw a chemical monkey wrench into its machinery. The proteins are folded globular structures with precisely defined functions and their “active sites” can be targeted with chemical compounds.
The first step is to clone and express the genes of the virus proteins and grow them as protein crystals in miniature ice cube-like trays. The consortium includes nine labs across eight institutions that will participate in this effort.
Above is a modified version of the Northwestern University news release written by Marla Paul.
Note: This will be an ongoing curation as new information and tools become available.
RNASeq is a powerful tool for the analysis of the transcriptome profile and has been used to determine the transcriptional changes occurring upon stimuli such as drug treatment or detecting transcript differences between biological sample cohorts such as tumor versus normal tissue. Unlike its genomic companion, whole genome and whole exome sequencing, which analyzes the primary sequence of the genomic DNA, RNASeq analyzes the mRNA transcripts, thereby more closely resembling the ultimate translated proteome. In addition, RNASeq and transcriptome profiling can determine if splicing variants occur as well as determining the nonexomic sequences, such as miRNA and lncRNA species, all of which have shown pertinence in the etiology of many diseases, including cancer.
However, RNASeq, like other omic technologies, generates enormous big data sets, which requires multiple types of bioinformatic tools in order to correctly analyze the sequence reads, and to visualize and interpret the output data. This post represents a curation by the RNA-Seq blog of such tools useful for RNASeq studies and lists and reviews published literature using these curated tools.
Core member and chair of the faculty, Broad Institute of MIT and Harvard; director, Klarman Cell Observatory, Broad Institute of MIT and Harvard; professor of biology, MIT; investigator, Howard Hughes Medical Institute; founding co-chair, Human Cell Atlas.
millions of genome variants, tens of thousands of disease-associated genes, thousands of cell types and an almost unimaginable number of ways they can combine, we had to approximate a best starting point—choose one target, guess the cell, simplify the experiment.
In 2020, advances in polygenic risk scores, in understanding the cell and modules of action of genes through genome-wide association studies (GWAS), and in predicting the impact of combinations of interventions.
we need algorithms to make better computational predictions of experiments we have never performed in the lab or in clinical trials.
Human Cell Atlas and the International Common Disease Alliance—and in new experimental platforms: data platforms and algorithms. But we also need a broader ecosystem of partnerships in medicine that engages interaction between clinical experts and mathematicians, computer scientists and engineers
Feng Zhang, PhD
investigator, Howard Hughes Medical Institute; core member, Broad Institute of MIT and Harvard; James and Patricia Poitras Professor of Neuroscience, McGovern Institute for Brain Research, MIT.
fundamental shift in medicine away from treating symptoms of disease and toward treating disease at its genetic roots.
Gene therapy with clinical feasibility, improved delivery methods and the development of robust molecular technologies for gene editing in human cells, affordable genome sequencing has accelerated our ability to identify the genetic causes of disease.
1,000 clinical trials testing gene therapies are ongoing, and the pace of clinical development is likely to accelerate.
refine molecular technologies for gene editing, to push our understanding of gene function in health and disease forward, and to engage with all members of society
Elizabeth Jaffee, PhD
Dana and Albert “Cubby” Broccoli Professor of Oncology, Johns Hopkins School of Medicine; deputy director, Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins.
a single blood test could inform individuals of the diseases they are at risk of (diabetes, cancer, heart disease, etc.) and that safe interventions will be available.
developing cancer vaccines. Vaccines targeting the causative agents of cervical and hepatocellular cancers have already proven to be effective. With these technologies and the wealth of data that will become available as precision medicine becomes more routine, new discoveries identifying the earliest genetic and inflammatory changes occurring within a cell as it transitions into a pre-cancer can be expected. With these discoveries, the opportunities to develop vaccine approaches preventing cancers development will grow.
shape how the culture of research will develop over the next 25 years, a culture that cares more about what is achieved than how it is achieved.
building a creative, inclusive and open research culture will unleash greater discoveries with greater impact.
John Nkengasong, PhD
Director, Africa Centres for Disease Control and Prevention.
To meet its health challenges by 2050, the continent will have to be innovative in order to leapfrog toward solutions in public health.
Precision medicine will need to take center stage in a new public health order— whereby a more precise and targeted approach to screening, diagnosis, treatment and, potentially, cure is based on each patient’s unique genetic and biologic make-up.
Eric Topol, MD
Executive vice-president, Scripps Research Institute; founder and director, Scripps Research Translational Institute.
In 2045, a planetary health infrastructure based on deep, longitudinal, multimodal human data, ideally collected from and accessible to as many as possible of the 9+ billion people projected to then inhabit the Earth.
enhanced capabilities to perform functions that are not feasible now.
AI machines’ ability to ingest and process biomedical text at scale—such as the corpus of the up-to-date medical literature—will be used routinely by physicians and patients.
the concept of a learning health system will be redefined by AI.
Linda Partridge, PhD
Professor, Max Planck Institute for Biology of Ageing.
Geroprotective drugs, which target the underlying molecular mechanisms of ageing, are coming over the scientific and clinical horizons, and may help to prevent the most intractable age-related disease, dementia.
Trevor Mundel, MD
President of Global Health, Bill & Melinda Gates Foundation.
finding new ways to share clinical data that are as open as possible and as closed as necessary.
moving beyond drug donations toward a new era of corporate social responsibility that encourages biotechnology and pharmaceutical companies to offer their best minds and their most promising platforms.
working with governments and multilateral organizations much earlier in the product life cycle to finance the introduction of new interventions and to ensure the sustainable development of the health systems that will deliver them.
deliver on the promise of global health equity.
Josep Tabernero, MD, PhD
Vall d’Hebron Institute of Oncology (VHIO); president, European Society for Medical Oncology (2018–2019).
genomic-driven analysis will continue to broaden the impact of personalized medicine in healthcare globally.
Precision medicine will continue to deliver its new paradigm in cancer care and reach more patients.
Immunotherapy will deliver on its promise to dismantle cancer’s armory across tumor types.
AI will help guide the development of individually matched
genetic patient screenings
the promise of liquid biopsy policing of disease?
Pardis Sabeti, PhD
Professor, Harvard University & Harvard T.H. Chan School of Public Health and Broad Institute of MIT and Harvard; investigator, Howard Hughes Medical Institute.
the development and integration of tools into an early-warning system embedded into healthcare systems around the world could revolutionize infectious disease detection and response.
But this will only happen with a commitment from the global community.
Els Toreele, PhD
Executive director, Médecins Sans Frontières Access Campaign
we need a paradigm shift such that medicines are no longer lucrative market commodities but are global public health goods—available to all those who need them.
This will require members of the scientific community to go beyond their role as researchers and actively engage in R&D policy reform mandating health research in the public interest and ensuring that the results of their work benefit many more people.
The global research community can lead the way toward public-interest driven health innovation, by undertaking collaborative open science and piloting not-for-profit R&D strategies that positively impact people’s lives globally.
Evolution of the Human Cell Genome Biology Field of Gene Expression, Gene Regulation, Gene Regulatory Networks and Application of Machine Learning Algorithms in Large-Scale Biological Data Analysis
How transcription factors (TFs) interpret cis-regulatory DNA sequence to control gene expression remains unclear, largely because past studies using native and engineered sequences had insufficient scale. Here, we measure the expression output of >100 million synthetic yeast promoter sequences that are fully random. These sequences yield diverse, reproducible expression levels that can be explained by their chance inclusion of functional TF binding sites. We use machine learning to build interpretable models of transcriptional regulation that predict ~94% of the expression driven from independent test promoters and ~89% of the expression driven from native yeast promoter fragments. These models allow us to characterize each TF’s specificity, activity and interactions with chromatin. TF activity depends on binding-site strand, position, DNA helical face and chromatin context. Notably, expression level is influenced by weak regulatory interactions, which confound designed-sequence studies. Our analyses show that massive-throughput assays of fully random DNA can provide the big data necessary to develop complex, predictive models of gene regulation.
The Evolution of the Human Cell Genome Biology Field of Gene Expression, Gene Regulation, Gene Regulatory Networks and Application of Machine Learning Algorithms in Large-Scale Biological Data Analysis is presented in the following Table
50
Liu, X., Li, Y. I. & Pritchard, J. K. Trans effects on gene expression can drive omnigenic inheritance. Cell 177, 1022–1034 e1026 (2019).
5
Muerdter, F. et al. Resolving systematic errors in widely used enhancer activity assays in human cells. Nat. Methods 15, 141–149 (2018).
6
Wang, X. et al. High-resolution genome-wide functional dissection of transcriptional regulatory regions and nucleotides in human. Nat. Commun. 9, 5380 (2018).
15
Yona, A. H., Alm, E. J. & Gore, J. Random sequences rapidly evolve into de novo promoters. Nat. Commun. 9, 1530 (2018).
4
van Arensbergen, J. et al. Genome-wide mapping of autonomous promoter activity in human cells. Nat. Biotechnol. 35, 145–153 (2017).
14
Cuperus, J. T. et al. Deep learning of the regulatory grammar of yeast 5’ untranslated regions from 500,000 random sequences. Genome Res. 27, 2015–2024 (2017).
31
Levo, M. et al. Systematic investigation of transcription factor activity in the context of chromatin using massively parallel binding and expression assays. Mol. Cell 65, 604–617 e606 (2017).
49
Boyle, E. A., Li, Y. I. & Pritchard, J. K. An expanded view of complex traits: from polygenic to omnigenic. Cell 169, 1177–1186 (2017).
Abadi, M. et al. TensorFlow: large-scale machine learning on heterogeneous systems. arXiv 1603.04467 (2016).
20
Shalem, O. et al. Systematic dissection of the sequence determinants of gene 3’ end mediated expression control. PLoS Genet. 11, e1005147 (2015).
55
Deng, C., Daley, T. & Smith, A. D. Applications of species accumulation curves in large-scale biological data analysis. Quant. Biol. 3, 135–144 (2015).
9
Hughes, T. R. & de Boer, C. G. Mapping yeast transcriptional networks. Genetics 195, 9–36 (2013).
10
Jolma, A. et al. DNA-binding specificities of human transcription factors. Cell 152, 327–339 (2013).
19
Kosuri, S. et al. Composability of regulatory sequences controlling transcription and translation in Escherichia coli. Proc. Natl Acad. Sci. USA 110, 14024–14029 (2013).
7
Sharon, E. et al. Inferring gene regulatory logic from high-throughput measurements of thousands of systematically designed promoters. Nat. Biotechnol. 30, 521–530 (2012).
18
de Boer, C. G. & Hughes, T. R. YeTFaSCo: a database of evaluated yeast transcription factor sequence specificities. Nucleic Acids Res. 40, D169–D179 (2012).
56
Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).
61
Cherry, J. M. et al. Saccharomyces Genome Database: the genomics resource of budding yeast. Nucleic Acids Res. 40, D700–D705 (2012).
11
Nutiu, R. et al. Direct measurement of DNA affinity landscapes on a high-throughput sequencing instrument. Nat. Biotechnol. 29, 659–664 (2011).
26
Zhang, Z. et al. A packing mechanism for nucleosome organization reconstituted across a eukaryotic genome. Science 332, 977–980 (2011).
30
Ganapathi, M. et al. Extensive role of the general regulatory factors, Abf1 and Rap1, in determining genome-wide chromatin structure in budding yeast. Nucleic Acids Res. 39, 2032–2044 (2011).
52
Erb, I. & van Nimwegen, E. Transcription factor binding site positioning in yeast: proximal promoter motifs characterize TATA-less promoters. PloS One 6, e24279 (2011).
3
Kinney, J. B., Murugan, A., Callan, C. G. Jr. & Cox, E. C. Using deep sequencing to characterize the biophysical mechanism of a transcriptional regulatory sequence. Proc. Natl Acad. Sci. USA107, 9158–9163 (2010).
8
Gertz, J., Siggia, E. D. & Cohen, B. A. Analysis of combinatorial cis-regulation in synthetic and genomic promoters. Nature 457, 215–218 (2009).
16
Wunderlich, Z. & Mirny, L. A. Different gene regulation strategies revealed by analysis of binding motifs. Trends Genet. 25, 434–440 (2009).
27
Hesselberth, J. R. et al. Global mapping of protein–DNA interactions in vivo by digital genomic footprinting. Nat. Methods 6, 283–289 (2009).
29
Hartley, P. D. & Madhani, H. D. Mechanisms that specify promoter nucleosome location and identity. Cell 137, 445–458 (2009).
51
Gibson, D. G. et al. Enzymatic assembly of DNA molecules up to several hundred kilobases. Nat. Methods 6, 343–345 (2009).
58
Segal, E. & Widom, J. From DNA sequence to transcriptional behaviour: a quantitative approach. Nat. Rev. Genet. 10, 443–456 (2009).
2
Yuan, Y., Guo, L., Shen, L. & Liu, J. S. Predicting gene expression from sequence: a reexamination. PLoS Comput. Biol. 3, e243 (2007).
46
Hibbs, M. A. et al. Exploring the functional landscape of gene expression: directed search of large microarray compendia. Bioinformatics 23, 2692–2699 (2007).
25
Liu, X., Lee, C. K., Granek, J. A., Clarke, N. D. & Lieb, J. D. Whole-genome comparison of Leu3 binding in vitro and in vivo reveals the importance of nucleosome occupancy in target site selection. Genome Res. 16, 1517–1528 (2006).
34
Roberts, G. G. & Hudson, A. P. Transcriptome profiling of Saccharomyces cerevisiae during a transition from fermentative to glycerol-based respiratory growth reveals extensive metabolic and structural remodeling. Mol. Genet. Genomics 276, 170–186 (2006).
48
Tanay, A. Extensive low-affinity transcriptional interactions in the yeast genome. Gen. Res. 16, 962–972 (2006).
53
Tong, A. H. & Boone, C. Synthetic genetic array analysis in Saccharomyces cerevisiae. Methods Mol. Biol. 313, 171–192 (2006).
57
Li, W. & Godzik, A. Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics 22, 1658–1659 (2006).
62
Chua, G. et al. Identifying transcription factor functions and targets by phenotypic activation. Proc. Natl Acad. Sci. USA 103, 12045–12050 (2006).
17
Arnosti, D. N. & Kulkarni, M. M. Transcriptional enhancers: intelligent enhanceosomes or flexible billboards? J. Cell. Biochem. 94, 890–898 (2005).
21
Granek, J. A. & Clarke, N. D. Explicit equilibrium modeling of transcription-factor binding and gene regulation. Genome Biol. 6, R87 (2005).
1
Beer, M. A. & Tavazoie, S. Predicting gene expression from sequence. Cell 117, 185–198 (2004).
28
Bernstein, B. E., Liu, C. L., Humphrey, E. L., Perlstein, E. O. & Schreiber, S. L. Global nucleosome occupancy in yeast. Genome Biol. 5, R62 (2004).
44
Kim, T. S., Kim, H. Y., Yoon, J. H. & Kang, H. S. Recruitment of the Swi/Snf complex by Ste12-Tec1 promotes Flo8-Mss11-mediated activation of STA1 expression. Mol. Cell. Biol. 24, 9542–9556 (2004).
45
Harbison, C. T. et al. Transcriptional regulatory code of a eukaryotic genome. Nature 431, 99–104 (2004).
60
Kent, N. A., Eibert, S. M. & Mellor, J. Cbf1p is required for chromatin remodeling at promoter-proximal CACGTG motifs in yeast. J. Biol. Chem. 279, 27116–27123 (2004).
22
Kulkarni, M. M. & Arnosti, D. N. Information display by transcriptional enhancers. Development 130, 6569–6575 (2003).
24
Conlon, E. M., Liu, X. S., Lieb, J. D. & Liu, J. S. Integrating regulatory motif discovery and genome-wide expression analysis. Proc. Natl Acad. Sci. USA 100, 3339–3344 (2003).
43
Neely, K. E., Hassan, A. H., Brown, C. E., Howe, L. & Workman, J. L. Transcription activator interactions with multiple SWI/SNF subunits. Mol. Cell. Biol. 22, 1615–1625 (2002).
23
Bussemaker, H. J., Li, H. & Siggia, E. D. Regulatory element detection using correlation with expression. Nat. Genet. 27, 167–171 (2001).
37
Haurie, V. et al. The transcriptional activator Cat8p provides a major contribution to the reprogramming of carbon metabolism during the diauxic shift in Saccharomyces cerevisiae. J. Biol. Chem. 276, 76–85 (2001).
39
Grauslund, M. & Ronnow, B. Carbon source-dependent transcriptional regulation of the mitochondrial glycerol-3-phosphate dehydrogenase gene, GUT2, from Saccharomyces cerevisiae. Can. J. Microbiol. 46, 1096–1100 (2000).
42
Cullen, P. J. & Sprague, G. F. Jr. Glucose depletion causes haploid invasive growth in yeast. Proc. Natl Acad. Sci. USA 97, 13619–13624 (2000).
38
Sato, T. et al. TheE-box DNA binding protein Sgc1p suppresses the gcr2 mutation, which is involved in transcriptional activation of glycolytic genes in Saccharomyces cerevisiae. FEBS Lett. 463, 307–311 (1999).
40
Madhani, H. D. & Fink, G. R. Combinatorial control required for the specificity of yeast MAPK signaling. Science 275, 1314–1317 (1997).
41
Gavrias, V., Andrianopoulos, A., Gimeno, C. J. & Timberlake, W. E. Saccharomyces cerevisiae TEC1 is required for pseudohyphal growth. Mol. Microbiol. 19, 1255–1263 (1996).
36
Hedges, D., Proft, M. & Entian, K. D. CAT8, a new zinc cluster-encoding gene necessary for derepression of gluconeogenic enzymes in the yeast Saccharomyces cerevisiae. Mol. Cell. Biol. 15, 1915–1922 (1995).
47
Bednar, J. et al. Determination of DNA persistence length by cryo-electron microscopy. Separation of the static and dynamic contributions to the apparent persistence length of DNA. J. Mol. Biol. 254, 579–594 (1995).
32
Axelrod, J. D., Reagan, M. S. & Majors, J. GAL4 disrupts a repressing nucleosome during activation of GAL1 transcription in vivo. Genes Dev. 7, 857–869 (1993).
33
Morse, R. H. Nucleosome disruption by transcription factor binding in yeast. Science 262, 1563–1566 (1993).
12
Oliphant, A. R., Brandl, C. J. & Struhl, K. Defining the sequence specificity of DNA-binding proteins by selecting binding sites from random-sequence oligonucleotides: analysis of yeast GCN4 protein. Mol. Cell. Biol. 9, 2944–2949 (1989).
35
Forsburg, S. L. & Guarente, L. Identification and characterization of HAP4: a third component of the CCAAT-bound HAP2/HAP3 heteromer. Genes Dev. 3, 1166–1178 (1989).
13
Horwitz, M. S. & Loeb, L. A. Promoters selected from random DNA sequences. Proc. Natl Acad. Sci. USA 83, 7405–7409 (1986).
To access each reference as a live link, go to the number in the first column in the Table and look it up in the List of References in the Link, below
Klarman Cell Observatory, Broad Institute of MIT and Harvard, Cambridge, MA, USA
Carl G. de Boer
, Eeshit Dhaval Vaishnav
, Nir Friedman
& Aviv Regev
Howard Hughes Medical Institute and Koch Institute of Integrative Cancer Research, Department of Biology, Massachusetts Institute of Technology, Cambridge, MA, USA
Eeshit Dhaval Vaishnav
& Aviv Regev
School of Computer Science and Institute of Life Sciences, The Hebrew University of Jerusalem, Jerusalem, Israel
Ronen Sadeh
& Nir Friedman
Initiative for Maximizing Student Development Program, University of New Mexico, Albuquerque, NM, USA
Esteban Luis Abeyta
Contributions
C.G.D. and A.R. drafted the manuscript, with all authors contributing. C.G.D. analyzed the data. C.G.D., E.D.V., E.L.A. and R.S. performed the experiments. A.R. and N.F. supervised the research.
A.R. is an SAB member of Thermo Fisher Scientific, Neogene Therapeutics, Asimov, and Syros Pharmaceuticals, an equity holder of Immunitas, and a founder of and equity holder in Celsius Therapeutics. All other authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Cite this article
Boer, C.G., Vaishnav, E.D., Sadeh, R. et al. Deciphering eukaryotic gene-regulatory logic with 100 million random promoters. Nat Biotechnol (2019) doi:10.1038/s41587-019-0315-8
Deep Learning extracts Histopathological Patterns and accurately discriminates 28 Cancer and 14 Normal Tissue Types: Pan-cancer Computational Histopathology Analysis
Reporter: Aviva Lev-Ari, PhD, RN
3.5.1.1 Deep Learning extracts Histopathological Patterns and accurately discriminates 28 Cancer and 14 Normal Tissue Types: Pan-cancer Computational Histopathology Analysis, Volume 2 (Volume Two: Latest in Genomics Methodologies for Therapeutics: Gene Editing, NGS and BioInformatics, Simulations and the Genome Ontology), Part 3: AI in Medicine
Pan-cancer computational histopathology reveals mutations, tumor composition and prognosis
Yu Fu1, Alexander W Jung1, Ramon Viñas Torne1, Santiago Gonzalez1,2, Harald Vöhringer1, Mercedes Jimenez-Linan3, Luiza Moore3,4, and Moritz Gerstung#1,5 # to whom correspondence should be addressed 1) European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Hinxton, UK. 2) Current affiliation: Institute for Research in Biomedicine (IRB Barcelona), Parc Científic de Barcelona, Barcelona, Spain. 3) Department of Pathology, Addenbrooke’s Hospital, Cambridge, UK. 4) Wellcome Sanger Institute, Hinxton, UK 5) European Molecular Biology Laboratory, Genome Biology Unit, Heidelberg, Germany.
Correspondence:
Dr Moritz Gerstung European Molecular Biology Laboratory European Bioinformatics Institute (EMBL-EBI) Hinxton, CB10 1SA UK. Tel: +44 (0) 1223 494636 E-mail: moritz.gerstung@ebi.ac.uk
Abstract
Pan-cancer computational histopathology reveals mutations, tumor composition and prognosis
Here we use deep transfer learning to quantify histopathological patterns across 17,396 H&E stained histopathology image slides from 28 cancer types and correlate these with underlying genomic and transcriptomic data. Pan-cancer computational histopathology (PC-CHiP) classifies the tissue origin across organ sites and provides highly accurate, spatially resolved tumor and normal distinction within a given slide. The learned computational histopathological features correlate with a large range of recurrent genetic aberrations, including whole genome duplications (WGDs), arm-level copy number gains and losses, focal amplifications and deletions as well as driver gene mutations within a range of cancer types. WGDs can be predicted in 25/27 cancer types (mean AUC=0.79) including those that were not part of model training. Similarly, we observe associations with 25% of mRNA transcript levels, which enables to learn and localise histopathological patterns of molecularly defined cell types on each slide. Lastly, we find that computational histopathology provides prognostic information augmenting histopathological subtyping and grading in the majority of cancers assessed, which pinpoints prognostically relevant areas such as necrosis or infiltrating lymphocytes on each tumour section. Taken together, these findings highlight the large potential of PC-CHiP to discover new molecular and prognostic associations, which can augment diagnostic workflows and lay out a rationale for integrating molecular and histopathological data.
● Pan-cancer computational histopathology analysis with deep learning extracts histopathological patterns and accurately discriminates 28 cancer and 14 normal tissue types
● Computational histopathology predicts whole genome duplications, focal amplifications and deletions, as well as driver gene mutations
● Wide-spread correlations with gene expression indicative of immune infiltration and proliferation
● Prognostic information augments conventional grading and histopathology subtyping in the majority of cancers
Discussion
Here we presented PC-CHiP, a pan-cancer transfer learning approach to extract computational histopathological features across 42 cancer and normal tissue types and their genomic, molecular and prognostic associations. Histopathological features, originally derived to classify different tissues, contained rich histologic and morphological signals predictive of a range of genomic and transcriptomic changes as well as survival. This shows that computer vision not only has the capacity to highly accurately reproduce predefined tissue labels, but also that this quantifies diverse histological patterns, which are predictive of a broad range of genomic and molecular traits, which were not part of the original training task. As the predictions are exclusively based on standard H&E-stained tissue sections, our analysis highlights the high potential of computational histopathology to digitally augment existing histopathological workflows. The strongest genomic associations were found for whole genome duplications, which can in part be explained by nuclear enlargement and increased nuclear intensities, but seemingly also stems from tumour grade and other histomorphological patterns contained in the high-dimensional computational histopathological features. Further, we observed associations with a range of chromosomal gains and losses, focal deletions and amplifications as well as driver gene mutations across a number of cancer types. These data demonstrate that genomic alterations change the morphology of cancer cells, as in the case of WGD, but possibly also that certain aberrations preferentially occur in distinct cell types, reflected by the tumor histology. Whatever is the cause or consequence in this equation, these associations lay out a route towards genomically defined histopathology subtypes, which will enhance and refine conventional assessment. Further, a broad range of transcriptomic correlations was observed reflecting both immune cell infiltration and cell proliferation that leads to higher tumor densities. These examples illustrated the remarkable property that machine learning does not only establish novel molecular associations from pre-computed histopathological feature sets but also allows the localisation of these traits within a larger image. While this exemplifies the power of a large scale data analysis to detect and localise recurrent patterns, it is probably not superior to spatially annotated training data. Yet such data can, by definition, only be generated for associations which are known beforehand. This appears straightforward, albeit laborious, for existing histopathology classifications, but more challenging for molecular readouts. Yet novel spatial transcriptomic44,45 and sequencing technologies46 bring within reach spatially matched molecular and histopathological data, which would serve as a gold standard in combining imaging and molecular patterns. Across cancer types, computational histopathological features showed a good level of prognostic relevance, substantially improving prognostic accuracy over conventional grading and histopathological subtyping in the majority of cancers. It is this very remarkable that such predictive It is made available under a CC-BY-NC 4.0 International license. (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. bioRxiv preprint first posted online Oct. 25, 2019; doi: http://dx.doi.org/10.1101/813543. The copyright holder for this preprint signals can be learned in a fully automated fashion. Still, at least at the current resolution, the improvement over a full molecular and clinical workup was relatively small. This might be a consequence of the far-ranging relations between histopathology and molecular phenotypes described here, implying that histopathology is a reflection of the underlying molecular alterations rather than an independent trait. Yet it probably also highlights the challenges of unambiguously quantifying histopathological signals in – and combining signals from – individual areas, which requires very large training datasets for each tumour entity. From a methodological point of view, the prediction of molecular traits can clearly be improved. In this analysis, we adopted – for the reason of simplicity and to avoid overfitting – a transfer learning approach in which an existing deep convolutional neural network, developed for classification of everyday objects, was fine tuned to predict cancer and normal tissue types. The implicit imaging feature representation was then used to predict molecular traits and outcomes. Instead of employing this two-step procedure, which risks missing patterns irrelevant for the initial classification task, one might directly employ either training on the molecular trait of interest, or ideally multi-objective learning. Further improvement may also be related to the choice of the CNN architecture. Everyday images have no defined scale due to a variable z-dimension; therefore, the algorithms need to be able to detect the same object at different sizes. This clearly is not the case for histopathology slides, in which one pixel corresponds to a defined physical size at a given magnification. Therefore, possibly less complex CNN architectures may be sufficient for quantitative histopathology analyses, and also show better generalisation. Here, in our proof-of-concept analysis, we observed a considerable dependence of the feature representation on known and possibly unknown properties of our training data, including the image compression algorithm and its parameters. Some of these issues could be overcome by amending and retraining the network to isolate the effect of confounding factors and additional data augmentation. Still, given the flexibility of deep learning algorithms and the associated risk of overfitting, one should generally be cautious about the generalisation properties and critically assess whether a new image is appropriately represented. Looking forward, our analyses revealed the enormous potential of using computer vision alongside molecular profiling. While the eye of a trained human may still constitute the gold standard for recognising clinically relevant histopathological patterns, computers have the capacity to augment this process by sifting through millions of images to retrieve similar patterns and establish associations with known and novel traits. As our analysis showed this helps to detect histopathology patterns associated with a range of genomic alterations, transcriptional signatures and prognosis – and highlight areas indicative of these traits on each given slide. It is therefore not too difficult to foresee how this may be utilised in a computationally augmented histopathology workflow enabling more precise and faster diagnosis and prognosis. Further, the ability to quantify a rich set of histopathology patterns lays out a path to define integrated histopathology and molecular cancer subtypes, as recently demonstrated for colorectal cancers47 .
Lastly, our analyses provide It is made available under a CC-BY-NC 4.0 International license. (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint proof-of-concept for these principles and we expect them to be greatly refined in the future based on larger training corpora and further algorithmic refinements.
Other related articles published in this Open Access Online Scientific Journal include the following:
CancerBase.org – The Global HUB for Diagnoses, Genomes, Pathology Images: A Real-time Diagnosis and Therapy Mapping Service for Cancer Patients – Anonymized Medical Records accessible to anyone on Earth
Single-cell RNA-seq helps in finding intra-tumoral heterogeneity in pancreatic cancer
Reporter and Curator: Dr. Sudipta Saha, Ph.D.
4.3.6 Single-cell RNA-seq helps in finding intra-tumoral heterogeneity in pancreatic cancer, Volume 2 (Volume Two: Latest in Genomics Methodologies for Therapeutics: Gene Editing, NGS and BioInformatics, Simulations and the Genome Ontology), Part 4: Single Cell Genomics
Pancreatic cancer is a significant cause of cancer mortality; therefore, the development of early diagnostic strategies and effective treatment is essential. Improvements in imaging technology, as well as use of biomarkers are changing the way that pancreas cancer is diagnosed and staged. Although progress in treatment for pancreas cancer has been incremental, development of combination therapies involving both chemotherapeutic and biologic agents is ongoing.
Cancer is an evolutionary disease, containing the hallmarks of an asexually reproducing unicellular organism subject to evolutionary paradigms. Pancreatic ductal adenocarcinoma (PDAC) is a particularly robust example of this phenomenon. Genomic features indicate that pancreatic cancer cells are selected for fitness advantages when encountering the geographic and resource-depleted constraints of the microenvironment. Phenotypic adaptations to these pressures help disseminated cells to survive in secondary sites, a major clinical problem for patients with this disease.
The immune system varies in cell types, states, and locations. The complex networks, interactions, and responses of immune cells produce diverse cellular ecosystems composed of multiple cell types, accompanied by genetic diversity in antigen receptors. Within this ecosystem, innate and adaptive immune cells maintain and protect tissue function, integrity, and homeostasis upon changes in functional demands and diverse insults. Characterizing this inherent complexity requires studies at single-cell resolution. Recent advances such as massively parallel single-cell RNA sequencing and sophisticated computational methods are catalyzing a revolution in our understanding of immunology.
PDAC is the most common type of pancreatic cancer featured with high intra-tumoral heterogeneity and poor prognosis. In the present study to comprehensively delineate the PDAC intra-tumoral heterogeneity and the underlying mechanism for PDAC progression, single-cell RNA-seq (scRNA-seq) was employed to acquire the transcriptomic atlas of 57,530 individual pancreatic cells from primary PDAC tumors and control pancreases. The diverse malignant and stromal cell types, including two ductal subtypes with abnormal and malignant gene expression profiles respectively, were identified in PDAC.
The researchers found that the heterogenous malignant subtype was composed of several subpopulations with differential proliferative and migratory potentials. Cell trajectory analysis revealed that components of multiple tumor-related pathways and transcription factors (TFs) were differentially expressed along PDAC progression. Furthermore, it was found a subset of ductal cells with unique proliferative features were associated with an inactivation state in tumor-infiltrating T cells, providing novel markers for the prediction of antitumor immune response. Together, the findings provided a valuable resource for deciphering the intra-tumoral heterogeneity in PDAC and uncover a connection between tumor intrinsic transcriptional state and T cell activation, suggesting potential biomarkers for anticancer treatment such as targeted therapy and immunotherapy.
scPopCorn: A New Computational Method for Subpopulation Detection and their Comparative Analysis Across Single-Cell Experiments
Reporter and Curator: Dr. Sudipta Saha, Ph.D.
4.2.5 scPopCorn: A New Computational Method for Subpopulation Detection and their Comparative Analysis Across Single-Cell Experiments, Volume 2 (Volume Two: Latest in Genomics Methodologies for Therapeutics: Gene Editing, NGS and BioInformatics, Simulations and the Genome Ontology), Part 4: Single Cell Genomics
Present day technological advances have facilitated unprecedented opportunities for studying biological systems at single-cell level resolution. For example, single-cell RNA sequencing (scRNA-seq) enables the measurement of transcriptomic information of thousands of individual cells in one experiment. Analyses of such data provide information that was not accessible using bulk sequencing, which can only assess average properties of cell populations. Single-cell measurements, however, can capture the heterogeneity of a population of cells. In particular, single-cell studies allow for the identification of novel cell types, states, and dynamics.
One of the most prominent uses of the scRNA-seq technology is the identification of subpopulations of cells present in a sample and comparing such subpopulations across samples. Such information is crucial for understanding the heterogeneity of cells in a sample and for comparative analysis of samples from different conditions, tissues, and species. A frequently used approach is to cluster every dataset separately, inspect marker genes for each cluster, and compare these clusters in an attempt to determine which cell types were shared between samples. This approach, however, relies on the existence of predefined or clearly identifiable marker genes and their consistent measurement across subpopulations.
Although the aligned data can then be clustered to reveal subpopulations and their correspondence, solving the subpopulation-mapping problem by performing global alignment first and clustering second overlooks the original information about subpopulations existing in each experiment. In contrast, an approach addressing this problem directly might represent a more suitable solution. So, keeping this in mind the researchers developed a computational method, single-cell subpopulations comparison (scPopCorn), that allows for comparative analysis of two or more single-cell populations.
The performance of scPopCorn was tested in three distinct settings. First, its potential was demonstrated in identifying and aligning subpopulations from single-cell data from human and mouse pancreatic single-cell data. Next, scPopCorn was applied to the task of aligning biological replicates of mouse kidney single-cell data. scPopCorn achieved the best performance over the previously published tools. Finally, it was applied to compare populations of cells from cancer and healthy brain tissues, revealing the relation of neoplastic cells to neural cells and astrocytes. Consequently, as a result of this integrative approach, scPopCorn provides a powerful tool for comparative analysis of single-cell populations.
This scPopCorn is basically a computational method for the identification of subpopulations of cells present within individual single-cell experiments and mapping of these subpopulations across these experiments. Different from other approaches, scPopCorn performs the tasks of population identification and mapping simultaneously by optimizing a function that combines both objectives. When applied to complex biological data, scPopCorn outperforms previous methods. However, it should be kept in mind that scPopCorn assumes the input single-cell data to consist of separable subpopulations and it is not designed to perform a comparative analysis of single cell trajectories datasets that do not fulfill this constraint.
Several innovations developed in this work contributed to the performance of scPopCorn. First, unifying the above-mentioned tasks into a single problem statement allowed for integrating the signal from different experiments while identifying subpopulations within each experiment. Such an incorporation aids the reduction of biological and experimental noise. The researchers believe that the ideas introduced in scPopCorn not only enabled the design of a highly accurate identification of subpopulations and mapping approach, but can also provide a stepping stone for other tools to interrogate the relationships between single cell experiments.
Real Time @BIOConvention #BIO2019:#Bitcoin Your Data! From Trusted Pharma Silos to Trustless Community-Owned Blockchain-Based Precision Medicine Data Trials
As care for lifestyle-driven chronic diseases expands in scope, prevention and recovery are becoming the new areas of focus. Building a precision medicine foundation that will promote ownership of individuals’ health data and allow for sharing and trading of this data could prove a great blockchain.
At its core, blockchain may offer the potential of a shared platform that decentralizes healthcare interactions ensuring access control, authenticity and integrity, while presenting the industry with radical possibilities for value-based care and reimbursement models. Panelists will explore these new discoveries as well as look to answer lingering questions, such as: are we off to a “trustless” information model underpinned by Bitcoin cryptocurrency, where no central authority validates the transactions in the ledger, and anyone whose computers can do the required math can join to mine and add blocks to your data? Would smart contracts begin to incentivize “rational” behaviors where consumers respond in a manner that makes their data interesting?
Moderator: Cybersecurity is extremely important in the minds of healthcare CEOs. CEO of Kaiser Permenente has listed this as one of main concerns for his company.
Sanjeey of Singularity: There are Very few companies in this space. Singularity have collected thousands of patient data. They wanted to do predictive health care, where a patient will know beforehand what health problems and issues to expect. Created a program called Virtual Assistant. As data is dynamic, the goal was to provide Virtual Assistant to everyone.
Benefits of blockchain: secure, simple to update, decentralized data; patient can control their own data, who sees it and monetize it.
Nebular Genetics: Company was founded by Dr. George Church, who had pioneered the next generation sequencing (NGS) methodology. The company goal is to make genomics available to all but this currently is not the case as NGS is not being used as frequently.
The problem is a data problem:
data not organized
data too parsed
data not accessible
Blockchain may be able to alleviate the accessibiltiy problem. Pharma is very interested in the data but expensive to collect. In addition many companies just do large scale but low depth sequencing. For example 23andme (which had recently made a big deal with Lilly for data) only sequences about 1% of genome.
There are two types of genome sequencing companies
large scale and low depth – like 23andme
smaller scale but higher depth – like DECODE and some of the EU EXOME sequencing efforts like the 1000 Project
Simply Vital Health: Harnesses blockchain to combat ineffeciencies in hospital records. They tackle the costs after acute care so increase the value based care. Most of healthcare is concentrated on the top earners and little is concentrated on the majority less affluent and poor. On addressing HIPAA compliance issues: they decided to work with HIPAA and comply but will wait for this industry to catch up so the industry as a whole can lobby to affect policy change required for blockchain technology to work efficiently in this arena. They will only work with known vendors: VERY Important to know where the data is kept and who are controlling the servers you are using. With other blockchain like Etherium or Bitcoin, the servers are anonymous.
Encrypgen: generates new blockchain for genomic data and NGS companies.
Please follow LIVE on TWITTER using the following @ handles and # hashtags:
Real Time Coverage @BIOConvention #BIO2019: Issues of Risk and Reproduceability in Translational and Academic Collaboration; 2:30-4:00 June 3 Philadelphia PA
Reporter: Stephen J. Williams, PhD @StephenJWillia2Article ID #267: Real Time Coverage @BIOConvention #BIO2019: Issues of Risk and Reproduceability in Translational and Academic Collaboration; 2:30-4:00 June 3 Philadelphia PA. Published on 6/3/2019
WordCloud Image Produced by Adam Tubman
Translating academic research into products and new therapies is a very risky venture as only 1% of academic research has been successfully translated into successful products.
Collaboration from Chicago area universities like U of Chicago, Northwestern, etc. First phase was enhance collaboration between universities by funding faculty recruitment and basic research. Access to core facilities across universities. Have expanded to give alternatives to company formation.
Most academic PI are not as savvy to start a biotech so they bring in biotechs and build project teams as well as developing a team of ex pharma and biotech experts. Derisk as running as one asset project. Partner as early as possible. A third of their pipeline have been successfully partnered. Work with investors and patent attorneys.
Focused on getting PIs to get to startup. Focused on oncology and vaccines and I/O. The result can be liscensing or partnership. Running around 50 to 60 projects. Creating a new company from these US PI partnerships.
Most projects from Harvard have been therapeutics-based. At Harvard they have a network of investors ($50 million). They screen PI proposals based on translateability and what investors are interested in.
In Chicago they solicit multiple projects but are agnostic on area but as they are limited they are focused on projects that will assist in developing a stronger proposal to investor/funding mechanism.
NYU goes around university doing due diligence reaching out to investigators. They shop around their projects to wet their investors, pharma appetite future funding. At Takeda they have five centers around US. They want to have more input so go into the university with their scientists and discuss ideas.
Challenges:
Takeda: Data Validation very important. Second there may be disconnect with the amount of equity the PI wants in the new company as well as management. Third PIs not aware of all steps in drug development.
Harvard: Pharma and biotech have robust research and academic does not have the size or scope of pharma. PIs must be more diligent on e.g. the compounds they get from a screen… they only focus narrowly
NYU: bring in consultants as PIs don’t understand all the management issues. Need to understand development so they bring in the experts to help them. Pharma he feels have to much risk aversion and none of their PIs want 100% equity.
Chicago: they like to publish at early stage so publication freedom is a challenge
Dr. Freedman: Most scientists responding to Nature survey said yes a reproduceability crisis. The reasons: experimental bias, lack of validation techniques, reagents, and protocols etc.
And as he says there is a great ECONOMIC IMPACT of preclinical reproducability issues: to the tune of $56 billion of irreproducable results (paper published in PLOS Biology). If can find the core drivers of this issue they can solve the problem. STANDARDS are constantly used in various industries however academic research are lagging in developing such standards. Just the problem of cell line authentication is costing $4 billion.
Dr. Cousins: There are multiple high throughput screening (HTS) academic centers around the world (150 in US). So where does the industry go for best practices in assays? Eli Lilly had developed a manual for HTS best practices and in 1984 made publicly available (Assay Guidance Manual). To date there have been constant updates to this manual to incorporate new assays. Workshops have been developed to train scientists in these best practices.
NIH has been developing new programs to address these reproducability issues. Developed a method called
“Ring Testing Initiative” where multiple centers involved in sharing reagents as well as assays and allowing scientists to test at multiple facilities.
Dr.Tong: Reproduceability of Microarrays: As microarrays were the only methodology to do high through put genomics in the early 2000s, and although much research had been performed to standardize and achieve best reproduceability of the microarray technology (determining best practices in spotting RNA on glass slides, hybridization protocols, image analysis) little had been done on evaluating the reproducibility of results obtained from microarray experiments involving biological samples. The advent of Artificial Intelligence and Machine Learning though can be used to help validate microarray results. This was done in a Nature Biotechnology paper (Nature Biotechnologyvolume28, pages827–838 (2010)) by an international consortium, the International MAQC (Microarray Quality Control) Society and can be found here
However Dr. Tong feels there is much confusion in how we define reproduceability. Dr. Tong identified a few key points of data reproduceability:
Traceability: what are the practices and procedures from going from point A to point B (steps in a protocol or experimental design)
Repeatability: ability to repeat results within the same laboratory
Replicatablilty: ability to repeat results cross laboratory
Transferability: are the results validated across multiple platforms?
The panel then discussed the role of journals and funders to drive reproduceability in research. They felt that editors have been doing as much as they can do as they receive an end product (the paper) but all agreed funders need to do more to promote data validity, especially in requiring that systematic evaluation and validation of each step in protocols are performed.. There could be more training of PIs with respect to protocol and data validation.
Other Articles on Industry/Academic Research Partnerships and Translational Research on this Open Access Online Journal Include
BioInformatic Resources at the Environmental Protection Agency: Tools and Webinars on Toxicity Prediction, Volume 2 (Volume Two: Latest in Genomics Methodologies for Therapeutics: Gene Editing, NGS and BioInformatics, Simulations and the Genome Ontology), Part 1: Next Generation Sequencing (NGS)
BioInformatic Resources at the Environmental Protection Agency: Tools and Webinars on Toxicity Prediction
Curator Stephen J. Williams Ph.D.
New GenRA Module in EPA’s CompTox Dashboard Will Help Predict Potential Chemical Toxicity
Published September 25, 2018
As part of its ongoing computational toxicology research, EPA is developing faster and improved approaches to evaluate chemicals for potential health effects. One commonly applied approach is known as chemical read-across. Read-across uses information about how a chemical with known data behaves to make a prediction about the behavior of another chemical that is “similar” but does not have as much data. Current read-across, while cost-effective, relies on a subjective assessment, which leads to varying predictions and justifications depending on who undertakes and evaluates the assessment.
To reduce uncertainties and develop a more objective approach, EPA researchers have developed an automated read-across tool called Generalized Read-Across (GenRA), and added it to the newest version of the EPA Computational Toxicology Dashboard. The goal of GenRA is to encode as many expert considerations used within current read-across approaches as possible and combine these with data-driven approaches to transition read-across towards a more systematic and data-based method of making predictions.
EPA chemist Dr. Grace Patlewicz says it was this uncertainty that motivated the development of GenRA. “You don’t actually know if you’ve been successful at using read-across to help predict chemical toxicity because it’s a judgement call based on one person versus the next. That subjectivity is something we were trying to move away from.” Patlewicz says.
Since toxicologists and risk assessors are already familiar with read-across, EPA researchers saw value in creating a tool that that was aligned with the current read-across workflow but which addressed uncertainty using data analysis methods in what they call a “harmonized-hybrid workflow.”
In its current form, GenRA lets users find analogues, or chemicals that are similar to their target chemical, based on chemical structural similarity. The user can then select which analogues they want to carry forward into the GenRA prediction by exploring the consistency and concordance of the underlying experimental data for those analogues. Next, the tool predicts toxicity effects of specific repeated dose studies. Then, a plot with these outcomes is generated based on a similarity-weighted activity of the analogue chemicals the user selected. Finally, the user is presented with a data matrix view showing whether a chemical is predicted to be toxic (yes or no) for a chosen set of toxicity endpoints, with a quantitative measure of uncertainty.
The team is also comparing chemicals based on other similarity contexts, such as physicochemical characteristics or metabolic similarity, as well as extending the approach to make quantitative predictions of toxicity.
Patlewicz thinks incorporating other contexts and similarity measures will refine GenRA to make better toxicity predictions, fulfilling the goal of creating a read-across method capable of assessing thousands of chemicals that currently lack toxicity data.
“That’s the direction that we’re going in,” Patlewicz says. “Recognizing where we are and trying to move towards something a little bit more objective, showing how aspects of the current read-across workflow could be refined.”
Benchmark Dose Software (BMDS) EPA developed the Benchmark Dose Software (BMDS) as a tool to help estimate dose or exposure of a chemical or chemical mixture associated with a given response level. The methodology is used by EPA risk assessors and is fast becoming the world’s standard for dose-response analysis for risk assessments, including air pollution risk assessments.
BenMAP BenMAP is a Windows-based computer program that uses a Geographic Information System (GIS)-based to estimate the health impacts and economic benefits occurring when populations experience changes in air quality.
Community-Focused Exposure and Risk Screening Tool (C-FERST) C-FERST is an online tool developed by EPA in collaboration with stakeholders to provide access to resources that can be used with communities to help identify and learn more about their environmental health issues and explore exposure and risk reduction options.
Community Health Vulnerability Index EPA scientists developed a Community Health Vulnerability Index that can be used to help identify communities at higher health risk from wildfire smoke. Breathing smoke from a nearby wildfire is a health threat, especially for people with lung or heart disease, diabetes and high blood pressure as well as older adults, and those living in communities with poverty, unemployment and other indicators of social stress. Health officials can use the tool, in combination with air quality models, to focus public health strategies on vulnerable populations living in areas where air quality is impaired, either by wildfire smoke or other sources of pollution. The work was published in Environmental Science & Technology.
Critical Loads Mapper Tool The Critical Loads Mapper Tool can be used to help protect terrestrial and aquatic ecosystems from atmospheric deposition of nitrogen and sulfur, two pollutants emitted from fossil fuel burning and agricultural emissions. The interactive tool provides easy access to information on deposition levels through time; critical loads, which identify thresholds when pollutants have reached harmful levels; and exceedances of these thresholds.
EnviroAtlas EnviroAtlas provides interactive tools and resources for exploring the benefits people receive from nature or “ecosystem goods and services”. Ecosystem goods and services are critically important to human health and well-being, but they are often overlooked due to lack of information. Using EnviroAtlas, many types of users can access, view, and analyze diverse information to better understand the potential impacts of various decisions.
EPA Air Sensor Toolbox for Citizen Scientists EPA’s Air Sensor Toolbox for Citizen Scientists provides information and guidance on new low-cost compact technologies for measuring air quality. Citizens are interested in learning more about local air quality where they live, work and play. EPA’s Toolbox includes information about: Sampling methodologies; Calibration and validation approaches; Measurement methods options; Data interpretation guidelines; Education and outreach; and Low cost sensor performance information.
ExpoFIRST The Exposure Factors Interactive Resource for Scenarios Tool (ExpoFIRST) brings data from EPA’s Exposure Factors Handbook: 2011 Edition (EFH) to an interactive tool that maximizes flexibility and transparency for exposure assessors. ExpoFIRST represents a significant advance for regional, state, and local scientists in performing and documenting calculations for community and site-specific exposure assessments, including air pollution exposure assessments.
EXPOsure toolbox (ExpoBox) This is a toolbox created to assist individuals from within government, industry, academia, and the general public with assessing exposure, including exposure to air contaminants, fate and transport processes of air pollutants and their potential exposure concentrations. It is a compendium of exposure assessment tools that links to guidance documents, databases, models, reference materials, and other related resources.
Federal Reference & Federal Equivalency Methods EPA scientists develop and evaluate Federal Reference Methods and Federal Equivalency Methods for accurately and reliably measuring six primary air pollutants in outdoor air. These methods are used by states and other organizations to assess implementation actions needed to attain National Ambient Air Quality Standards.
Fertilizer Emission Scenario Tool for CMAQ (FEST-C) FEST-C facilitates the definition and simulation of new cropland farm management system scenarios or editing of existing scenarios to drive Environmental Policy Integrated Climate model (EPIC) simulations. For the standard 12km continental Community Multi-Scale Air Quality model (CMAQ) domain, this amounts to about 250,000 simulations for the U.S. alone. It also produces gridded daily EPIC weather input files from existing hourly Meteorology-Chemistry Interface Processor (MCIP) files, transforms EPIC output files to CMAQ-ready input files and links directly to Visual Environment for Rich Data Interpretation (VERDI) for spatial visualization of input and output files. The December 2012 release will perform all these functions for any CMAQ grid scale or domain.
Integrated Climate and Land use Scenarios (ICLUS) Climate change and land-use change are global drivers of environmental change. Impact assessments frequently show that interactions between climate and land-use changes can create serious challenges for aquatic ecosystems, water quality, and air quality. Population projections to 2100 were used to model the distribution of new housing across the landscape. In addition, housing density was used to estimate changes in impervious surface cover. A final report, datasets, the ICLUS+ Web Viewer and ArcGIS tools are available.
Indoor Semi-Volatile Organic Compound (i-SVOC) i-SVOC Version 1.0 is a general-purpose software application for dynamic modeling of the emission, transport, sorption, and distribution of semi-volatile organic compounds (SVOCs) in indoor environments. i-SVOC supports a variety of uses, including exposure assessment and the evaluation of mitigation options. SVOCs are a diverse group of organic chemicals that can be found in: Many are also present in indoor air, where they tend to bind to interior surfaces and particulate matter (dust).
Pesticides;
Ingredients in cleaning agents and personal care products;
Additives to vinyl flooring, furniture, clothing, cookware, food packaging, and electronics.
Municipal Solid Waste Decision Support Tool (MSW DST)EXIT This tool is designed to aid solid waste planners in evaluating the cost and environmental aspects of integrated municipal solid waste management strategies. The tool is the result of collaboration between EPA and RTI International and its partners.
Optical Noise-Reduction Averaging (ONA) Program Improves Black Carbon Particle Measurements Using Aethalometers ONA is a program that reduces noise in real-time black carbon data obtained using Aethalometers. Aethalometers optically measure the concentration of light absorbing or “black” particles that accumulate on a filter as air flows through it. These particles are produced by incomplete fossil fuel, biofuel and biomass combustion. Under polluted conditions, they appear as smoke or haze.
RETIGO tool Real Time Geospatial Data Viewer (RETIGO) is a free, web-based tool that shows air quality data that are collected while in motion (walking, biking or in a vehicle). The tool helps users overcome technical barriers to exploring air quality data. After collecting measurements, citizen scientists and other users can import their own data and explore the data on a map.
Remote Sensing Information Gateway (RSIG) RSIG offers a new way for users to get the multi-terabyte, environmental datasets they want via an interactive, Web browser-based application. A file download and parsing process that now takes months will be reduced via RSIG to minutes.
Simulation Tool Kit for Indoor Air Quality and Inhalation Exposure (IAQX) IAQX version 1.1 is an indoor air quality (IAQ) simulation software package that complements and supplements existing indoor air quality simulation (IAQ) programs. IAQX is for advanced users who have experience with exposure estimation, pollution control, risk assessment, and risk management. There are many sources of indoor air pollution, such as building materials, furnishings, and chemical cleaners. Since most people spend a large portion of their time indoors, it is important to be able to estimate exposure to these pollutants. IAQX helps users analyze the impact of pollutant sources and sinks, ventilation, and air cleaners. It performs conventional IAQ simulations to calculate the pollutant concentration and/or personal exposure as a function of time. It can also estimate adequate ventilation rates based on user-provided air quality criteria. This is a unique feature useful for product stewardship and risk management.
Spatial Allocator The Spatial Allocator provides tools that could be used by the air quality modeling community to perform commonly needed spatial tasks without requiring the use of a commercial Geographic Information System (GIS).
Traceability Protocol for Assay and Certification of Gaseous Calibration Standards This is used to certify calibration gases for ambient and continuous emission monitors. It specifies methods for assaying gases and establishing traceability to National Institute of Standards and Technology (NIST) reference standards. Traceability is required under EPA ambient and continuous emission monitoring regulations.
Watershed Deposition Mapping Tool (WDT) WDT provides an easy to use tool for mapping the deposition estimates from CMAQ to watersheds to provide the linkage of air and water needed for TMDL (Total Maximum Daily Load) and related nonpoint-source watershed analyses.
Visual Environment for Rich Data Interpretation (VERDI) VERDI is a flexible, modular, Java-based program for visualizing multivariate gridded meteorology, emissions, and air quality modeling data created by environmental modeling systems such as CMAQ and the Weather Research and Forecasting (WRF) model.
Databases
Air Quality Data for the CDC National Environmental Public Health Tracking Network EPA’s Exposure Research scientists are collaborating with the Centers for Disease Control and Prevention (CDC) on a CDC initiative to build a National Environmental Public Health Tracking (EPHT) network. Working with state, local and federal air pollution and health agencies, the EPHT program is facilitating the collection, integration, analysis, interpretation, and dissemination of data from environmental hazard monitoring, and from human exposure and health effects surveillance. These data provide scientific information to develop surveillance indicators, and to investigate possible relationships between environmental exposures, chronic disease, and other diseases, that can lead to interventions to reduce the burden of theses illnesses. An important part of the initiative is air quality modeling estimates and air quality monitoring data, combined through Bayesian modeling that can be linked with health outcome data.
EPAUS9R – An Energy Systems Database for use with the Market Allocation (MARKAL) Model The EPAUS9r is a regional database representation of the United States energy system. The database uses the MARKAL model. MARKAL is an energy system optimization model used by local and federal governments, national and international communities and academia. EPAUS9r represents energy supply, technology, and demand throughout the major sectors of the U.S. energy system.
Health & Environmental Research Online (HERO) HERO provides access to scientific literature used to support EPA’s integrated science assessments, including the Integrated Science Assessments (ISA) that feed into the National Ambient Air Quality (NAAQS) reviews.
SPECIATE 4.5 Database SPECIATE is a repository of volatile organic gas and particulate matter (PM) speciation profiles of air pollution sources.
A listing of EPA Tools and Databases for Water Contaminant Exposure Assessment
Exposure and Toxicity
EPA ExpoBox (A Toolbox for Exposure Assessors) This toolbox assists individuals from within government, industry, academia, and the general public with assessing exposure from multiple media, including water and sediment. It is a compendium of exposure assessment tools that links to guidance documents, databases, models, reference materials, and other related resources.
Chemical and Product Categories (CPCat) Database CPCat is a database containing information mapping more than 43,000 chemicals to a set of terms categorizing their usage or function. The comprehensive list of chemicals with associated categories of chemical and product use was compiled from publically available sources. Unique use category taxonomies from each source are mapped onto a single common set of approximately 800 terms. Users can search for chemicals by chemical name, Chemical Abstracts Registry Number, or by CPCat terms associated with chemicals.
A listing of EPA Tools and Databases for Chemical Toxicity Prediction & Assessment
Sequence Alignment to Predict Across Species Susceptibility (SeqAPASS) SeqAPASS is a fast, online screening tool that allows researchers and regulators to extrapolate toxicity information across species. For some species, such as humans, mice, rats, and zebrafish, the EPA has a large amount of data regarding their toxicological susceptibility to various chemicals. However, the toxicity data for numerous other plants and animals is very limited. SeqAPASS extrapolates from these data rich model organisms to thousands of other non-target species to evaluate their specific potential chemical susceptibility.
Aggregated Computational Toxicology Resource (ACToR) ACToR is an online warehouse of publicly available chemical toxicity data and can be used to find all publicly available data about potential chemical risks to human health and the environment. ACToR aggregates data from over 1000 public sources on over 500,000 environmental chemicals searchable by chemical name, other identifiers and by chemical structure.
Chemical and Product Category Database Contains information on how chemicals are used in consumer products. It includes information mapping over 43,000 chemicals to a set of terms categorizing their usage in consumer products or function.
Ecotoxicology Database (EcoTox) Source for finding single chemical toxicity data for aquatic life, terrestrial plants and wildlife.
Toxicity Reference Database (ToxRefDB) ToxRefDB contains thousands of animal toxicity studies on hundreds of chemicals from 30 years worth of animal toxicity studies.
A listing of EPA Webinar and Literature on Bioinformatic Tools and Projects
Comparative Bioinformatics Applications for Developmental Toxicology
Discuss how the US EPA/NCCT is trying to solve the problem of too many chemicals, too high cost, and too much biological uncertainty Discuss the solution the ToxCast Program is proposing; a data-rich system to screen, classify and rank chemicals for further evaluation
CHEMOINFORMATIC AND BIOINFORMATIC CHALLENGES AT THE US ENVIRONMENTAL PROTECTION AGENCY.
This presentation will provide an overview of both the scientific program and the regulatory activities related to computational toxicology. This presentation will provide an overview of both the scientific program and the regulatory activities related to computational toxicology.
How Can We Use Bioinformatics to Predict Which Agents Will Cause Birth Defects?
The availability of genomic sequences from a growing number of human and model organisms has provided an explosion of data, information, and knowledge regarding biological systems and disease processes. High-throughput technologies such as DNA and protein microarray biochips are now standard tools for probing the cellular state and determining important cellular behaviors at the genomic/proteomic levels. While these newer technologies are beginning to provide important information on cellular reactions to toxicant exposure (toxicogenomics), a major challenge that remains is the formulation of a strategy to integrate transcript, protein, metabolite, and toxicity data. This integration will require new concepts and tools in bioinformatics. The U.S. National Library of Medicine’s Pubmed site includes 19 million citations and abstracts and continues to grow. The BDSM team is now working on assembling the literature’s unstructured data into a structured database and linking it to BDSM within a system that can then be used for testing and generating new hypotheses. This effort will generate data bases of entities (such as genes, proteins, metabolites, gene ontology processes) linked to PubMed identifiers/abstracts and providing information on the relationships between them. The end result will be an online/standalone tool that will help researchers to focus on the papers most relevant to their query and uncover hidden connections and obvious information gaps.
ADVANCED PROTEOMICS AND BIOINFORMATICS TOOLS IN TOXICOLOGY RESEARCH: OVERCOMING CHALLENGES TO PROVIDE SIGNIFICANT RESULTS
This presentation specifically addresses the advantages and limitations of state of the art gel, protein arrays and peptide-based labeling proteomic approaches to assess the effects of a suite of model T4 inhibitors on the thyroid axis of Xenopus laevis.
Bioinformatic Integration of in vivo Data and Literature-based Gene Associations for Prioritization of Adverse Outcome Pathway Development
Adverse outcome pathways (AOPs) describe a sequence of events, beginning with a molecular initiating event (MIE), proceeding via key events (KEs), and culminating in an adverse outcome (AO). A challenge for use of AOPs in a safety evaluation context has been identification of MIEs and KEs relevant for AOs observed in regulatory toxicity studies. In this work, we implemented a bioinformatic approach that leverages mechanistic information in the literature and the AOs measured in regulatory toxicity studies to prioritize putative MIEs and/or early KEs for AOP development relevant to chemical safety evaluation. The US Environmental Protection Agency Toxicity Reference Database (ToxRefDB, v2.0) contains effect information for >1000 chemicals curated from >5000 studies or summaries from sources including data evaluation records from the US EPA Office of Pesticide Programs, the National Toxicology Program (NTP), peer-reviewed literature, and pharmaceutical preclinical studies. To increase ToxRefDB interoperability, endpoint and effect information were cross-referenced with codes from the United Medical Language System, which enabled mapping of in vivo pathological effects from ToxRefDB to PubMed (via Medical Subject Headings or MeSH). This enabled linkage to any resource that is also connected to PubMed or indexed with MeSH. A publicly available bioinformatic tool, the Entity-MeSH Co-occurrence Network (EMCON), uses multiple data sources and a measure of mutual information to identify genes most related to a MeSH term. Using EMCON, gene sets were generated for endpoints of toxicological relevance in ToxRefDB linking putative KEs and/or MIEs. The Comparative Toxicogenomics Database was used to further filter important associations. As a proof of concept, thyroid-related effects and their highly associated genes were examined, and demonstrated relevant MIEs and early KEs for AOPs to describe thyroid-related AOs. The ToxRefDB to gene mapping for thyroid resulted in >50 unique gene to chemical relationships. Integrated use of EMCON and ToxRefDB data provides a basis for rapid and robust putative AOP development, as well as a novel means to generate mechanistic hypotheses for specific chemicals. This abstract does not necessarily reflect U.S. EPA policy. Abstract and Poster for 2019 Society of Toxicology annual meeting in March 2019
Bioinformatic Integration of in vivo Data and Literature-based Gene Associations for Prioritization of Adverse Outcome Pathway Development
Adverse outcome pathways (AOPs) describe a sequence of events, beginning with a molecular initiating event (MIE), proceeding via key events (KEs), and culminating in an adverse outcome (AO). A challenge for use of AOPs in a safety evaluation context has been identification of MIEs and KEs relevant for AOs observed in regulatory toxicity studies. In this work, we implemented a bioinformatic approach that leverages mechanistic information in the literature and the AOs measured in regulatory toxicity studies to prioritize putative MIEs and/or early KEs for AOP development relevant to chemical safety evaluation. The US Environmental Protection Agency Toxicity Reference Database (ToxRefDB, v2.0) contains effect information for >1000 chemicals curated from >5000 studies or summaries from sources including data evaluation records from the US EPA Office of Pesticide Programs, the National Toxicology Program (NTP), peer-reviewed literature, and pharmaceutical preclinical studies. To increase ToxRefDB interoperability, endpoint and effect information were cross-referenced with codes from the United Medical Language System, which enabled mapping of in vivo pathological effects from ToxRefDB to PubMed (via Medical Subject Headings or MeSH). This enabled linkage to any resource that is also connected to PubMed or indexed with MeSH. A publicly available bioinformatic tool, the Entity-MeSH Co-occurrence Network (EMCON), uses multiple data sources and a measure of mutual information to identify genes most related to a MeSH term. Using EMCON, gene sets were generated for endpoints of toxicological relevance in ToxRefDB linking putative KEs and/or MIEs. The Comparative Toxicogenomics Database was used to further filter important associations. As a proof of concept, thyroid-related effects and their highly associated genes were examined, and demonstrated relevant MIEs and early KEs for AOPs to describe thyroid-related AOs. The ToxRefDB to gene mapping for thyroid resulted in >50 unique gene to chemical relationships. Integrated use of EMCON and ToxRefDB data provides a basis for rapid and robust putative AOP development, as well as a novel means to generate mechanistic hypotheses for specific chemicals. This abstract does not necessarily reflect U.S. EPA policy. Abstract and Poster for 2019 Society of Toxicology annual meeting in March 2019
Bioinformatic Integration of in vivo Data and Literature-based Gene Associations for Prioritization of Adverse Outcome Pathway Development
Adverse outcome pathways (AOPs) describe a sequence of events, beginning with a molecular initiating event (MIE), proceeding via key events (KEs), and culminating in an adverse outcome (AO). A challenge for use of AOPs in a safety evaluation context has been identification of MIEs and KEs relevant for AOs observed in regulatory toxicity studies. In this work, we implemented a bioinformatic approach that leverages mechanistic information in the literature and the AOs measured in regulatory toxicity studies to prioritize putative MIEs and/or early KEs for AOP development relevant to chemical safety evaluation. The US Environmental Protection Agency Toxicity Reference Database (ToxRefDB, v2.0) contains effect information for >1000 chemicals curated from >5000 studies or summaries from sources including data evaluation records from the US EPA Office of Pesticide Programs, the National Toxicology Program (NTP), peer-reviewed literature, and pharmaceutical preclinical studies. To increase ToxRefDB interoperability, endpoint and effect information were cross-referenced with codes from the United Medical Language System, which enabled mapping of in vivo pathological effects from ToxRefDB to PubMed (via Medical Subject Headings or MeSH). This enabled linkage to any resource that is also connected to PubMed or indexed with MeSH. A publicly available bioinformatic tool, the Entity-MeSH Co-occurrence Network (EMCON), uses multiple data sources and a measure of mutual information to identify genes most related to a MeSH term. Using EMCON, gene sets were generated for endpoints of toxicological relevance in ToxRefDB linking putative KEs and/or MIEs. The Comparative Toxicogenomics Database was used to further filter important associations. As a proof of concept, thyroid-related effects and their highly associated genes were examined, and demonstrated relevant MIEs and early KEs for AOPs to describe thyroid-related AOs. The ToxRefDB to gene mapping for thyroid resulted in >50 unique gene to chemical relationships. Integrated use of EMCON and ToxRefDB data provides a basis for rapid and robust putative AOP development, as well as a novel means to generate mechanistic hypotheses for specific chemicals. This abstract does not necessarily reflect U.S. EPA policy. Abstract and Poster for 2019 Society of Toxicology annual meeting in March 2019
Bioinformatic Integration of in vivo Data and Literature-based Gene Associations for Prioritization of Adverse Outcome Pathway Development
Adverse outcome pathways (AOPs) describe a sequence of events, beginning with a molecular initiating event (MIE), proceeding via key events (KEs), and culminating in an adverse outcome (AO). A challenge for use of AOPs in a safety evaluation context has been identification of MIEs and KEs relevant for AOs observed in regulatory toxicity studies. In this work, we implemented a bioinformatic approach that leverages mechanistic information in the literature and the AOs measured in regulatory toxicity studies to prioritize putative MIEs and/or early KEs for AOP development relevant to chemical safety evaluation. The US Environmental Protection Agency Toxicity Reference Database (ToxRefDB, v2.0) contains effect information for >1000 chemicals curated from >5000 studies or summaries from sources including data evaluation records from the US EPA Office of Pesticide Programs, the National Toxicology Program (NTP), peer-reviewed literature, and pharmaceutical preclinical studies. To increase ToxRefDB interoperability, endpoint and effect information were cross-referenced with codes from the United Medical Language System, which enabled mapping of in vivo pathological effects from ToxRefDB to PubMed (via Medical Subject Headings or MeSH). This enabled linkage to any resource that is also connected to PubMed or indexed with MeSH. A publicly available bioinformatic tool, the Entity-MeSH Co-occurrence Network (EMCON), uses multiple data sources and a measure of mutual information to identify genes most related to a MeSH term. Using EMCON, gene sets were generated for endpoints of toxicological relevance in ToxRefDB linking putative KEs and/or MIEs. The Comparative Toxicogenomics Database was used to further filter important associations. As a proof of concept, thyroid-related effects and their highly associated genes were examined, and demonstrated relevant MIEs and early KEs for AOPs to describe thyroid-related AOs. The ToxRefDB to gene mapping for thyroid resulted in >50 unique gene to chemical relationships. Integrated use of EMCON and ToxRefDB data provides a basis for rapid and robust putative AOP development, as well as a novel means to generate mechanistic hypotheses for specific chemicals. This abstract does not necessarily reflect U.S. EPA policy. Abstract and Poster for 2019 Society of Toxicology annual meeting in March 2019
Bioinformatic Integration of in vivo Data and Literature-based Gene Associations for Prioritization of Adverse Outcome Pathway Development
Adverse outcome pathways (AOPs) describe a sequence of events, beginning with a molecular initiating event (MIE), proceeding via key events (KEs), and culminating in an adverse outcome (AO). A challenge for use of AOPs in a safety evaluation context has been identification of MIEs and KEs relevant for AOs observed in regulatory toxicity studies. In this work, we implemented a bioinformatic approach that leverages mechanistic information in the literature and the AOs measured in regulatory toxicity studies to prioritize putative MIEs and/or early KEs for AOP development relevant to chemical safety evaluation. The US Environmental Protection Agency Toxicity Reference Database (ToxRefDB, v2.0) contains effect information for >1000 chemicals curated from >5000 studies or summaries from sources including data evaluation records from the US EPA Office of Pesticide Programs, the National Toxicology Program (NTP), peer-reviewed literature, and pharmaceutical preclinical studies. To increase ToxRefDB interoperability, endpoint and effect information were cross-referenced with codes from the United Medical Language System, which enabled mapping of in vivo pathological effects from ToxRefDB to PubMed (via Medical Subject Headings or MeSH). This enabled linkage to any resource that is also connected to PubMed or indexed with MeSH. A publicly available bioinformatic tool, the Entity-MeSH Co-occurrence Network (EMCON), uses multiple data sources and a measure of mutual information to identify genes most related to a MeSH term. Using EMCON, gene sets were generated for endpoints of toxicological relevance in ToxRefDB linking putative KEs and/or MIEs. The Comparative Toxicogenomics Database was used to further filter important associations. As a proof of concept, thyroid-related effects and their highly associated genes were examined, and demonstrated relevant MIEs and early KEs for AOPs to describe thyroid-related AOs. The ToxRefDB to gene mapping for thyroid resulted in >50 unique gene to chemical relationships. Integrated use of EMCON and ToxRefDB data provides a basis for rapid and robust putative AOP development, as well as a novel means to generate mechanistic hypotheses for specific chemicals. This abstract does not necessarily reflect U.S. EPA policy. Abstract and Poster for 2019 Society of Toxicology annual meeting in March 2019
Bioinformatic Integration of in vivo Data and Literature-based Gene Associations for Prioritization of Adverse Outcome Pathway Development
Adverse outcome pathways (AOPs) describe a sequence of events, beginning with a molecular initiating event (MIE), proceeding via key events (KEs), and culminating in an adverse outcome (AO). A challenge for use of AOPs in a safety evaluation context has been identification of MIEs and KEs relevant for AOs observed in regulatory toxicity studies. In this work, we implemented a bioinformatic approach that leverages mechanistic information in the literature and the AOs measured in regulatory toxicity studies to prioritize putative MIEs and/or early KEs for AOP development relevant to chemical safety evaluation. The US Environmental Protection Agency Toxicity Reference Database (ToxRefDB, v2.0) contains effect information for >1000 chemicals curated from >5000 studies or summaries from sources including data evaluation records from the US EPA Office of Pesticide Programs, the National Toxicology Program (NTP), peer-reviewed literature, and pharmaceutical preclinical studies. To increase ToxRefDB interoperability, endpoint and effect information were cross-referenced with codes from the United Medical Language System, which enabled mapping of in vivo pathological effects from ToxRefDB to PubMed (via Medical Subject Headings or MeSH). This enabled linkage to any resource that is also connected to PubMed or indexed with MeSH. A publicly available bioinformatic tool, the Entity-MeSH Co-occurrence Network (EMCON), uses multiple data sources and a measure of mutual information to identify genes most related to a MeSH term. Using EMCON, gene sets were generated for endpoints of toxicological relevance in ToxRefDB linking putative KEs and/or MIEs. The Comparative Toxicogenomics Database was used to further filter important associations. As a proof of concept, thyroid-related effects and their highly associated genes were examined, and demonstrated relevant MIEs and early KEs for AOPs to describe thyroid-related AOs. The ToxRefDB to gene mapping for thyroid resulted in >50 unique gene to chemical relationships. Integrated use of EMCON and ToxRefDB data provides a basis for rapid and robust putative AOP development, as well as a novel means to generate mechanistic hypotheses for specific chemicals. This abstract does not necessarily reflect U.S. EPA policy. Abstract and Poster for 2019 Society of Toxicology annual meeting in March 2019
Bioinformatic Integration of in vivo Data and Literature-based Gene Associations for Prioritization of Adverse Outcome Pathway Development
Adverse outcome pathways (AOPs) describe a sequence of events, beginning with a molecular initiating event (MIE), proceeding via key events (KEs), and culminating in an adverse outcome (AO). A challenge for use of AOPs in a safety evaluation context has been identification of MIEs and KEs relevant for AOs observed in regulatory toxicity studies. In this work, we implemented a bioinformatic approach that leverages mechanistic information in the literature and the AOs measured in regulatory toxicity studies to prioritize putative MIEs and/or early KEs for AOP development relevant to chemical safety evaluation. The US Environmental Protection Agency Toxicity Reference Database (ToxRefDB, v2.0) contains effect information for >1000 chemicals curated from >5000 studies or summaries from sources including data evaluation records from the US EPA Office of Pesticide Programs, the National Toxicology Program (NTP), peer-reviewed literature, and pharmaceutical preclinical studies. To increase ToxRefDB interoperability, endpoint and effect information were cross-referenced with codes from the United Medical Language System, which enabled mapping of in vivo pathological effects from ToxRefDB to PubMed (via Medical Subject Headings or MeSH). This enabled linkage to any resource that is also connected to PubMed or indexed with MeSH. A publicly available bioinformatic tool, the Entity-MeSH Co-occurrence Network (EMCON), uses multiple data sources and a measure of mutual information to identify genes most related to a MeSH term. Using EMCON, gene sets were generated for endpoints of toxicological relevance in ToxRefDB linking putative KEs and/or MIEs. The Comparative Toxicogenomics Database was used to further filter important associations. As a proof of concept, thyroid-related effects and their highly associated genes were examined, and demonstrated relevant MIEs and early KEs for AOPs to describe thyroid-related AOs. The ToxRefDB to gene mapping for thyroid resulted in >50 unique gene to chemical relationships. Integrated use of EMCON and ToxRefDB data provides a basis for rapid and robust putative AOP development, as well as a novel means to generate mechanistic hypotheses for specific chemicals. This abstract does not necessarily reflect U.S. EPA policy. Abstract and Poster for 2019 Society of Toxicology annual meeting in March 2019
Bioinformatic Integration of in vivo Data and Literature-based Gene Associations for Prioritization of Adverse Outcome Pathway Development
Adverse outcome pathways (AOPs) describe a sequence of events, beginning with a molecular initiating event (MIE), proceeding via key events (KEs), and culminating in an adverse outcome (AO). A challenge for use of AOPs in a safety evaluation context has been identification of MIEs and KEs relevant for AOs observed in regulatory toxicity studies. In this work, we implemented a bioinformatic approach that leverages mechanistic information in the literature and the AOs measured in regulatory toxicity studies to prioritize putative MIEs and/or early KEs for AOP development relevant to chemical safety evaluation. The US Environmental Protection Agency Toxicity Reference Database (ToxRefDB, v2.0) contains effect information for >1000 chemicals curated from >5000 studies or summaries from sources including data evaluation records from the US EPA Office of Pesticide Programs, the National Toxicology Program (NTP), peer-reviewed literature, and pharmaceutical preclinical studies. To increase ToxRefDB interoperability, endpoint and effect information were cross-referenced with codes from the United Medical Language System, which enabled mapping of in vivo pathological effects from ToxRefDB to PubMed (via Medical Subject Headings or MeSH). This enabled linkage to any resource that is also connected to PubMed or indexed with MeSH. A publicly available bioinformatic tool, the Entity-MeSH Co-occurrence Network (EMCON), uses multiple data sources and a measure of mutual information to identify genes most related to a MeSH term. Using EMCON, gene sets were generated for endpoints of toxicological relevance in ToxRefDB linking putative KEs and/or MIEs. The Comparative Toxicogenomics Database was used to further filter important associations. As a proof of concept, thyroid-related effects and their highly associated genes were examined, and demonstrated relevant MIEs and early KEs for AOPs to describe thyroid-related AOs. The ToxRefDB to gene mapping for thyroid resulted in >50 unique gene to chemical relationships. Integrated use of EMCON and ToxRefDB data provides a basis for rapid and robust putative AOP development, as well as a novel means to generate mechanistic hypotheses for specific chemicals. This abstract does not necessarily reflect U.S. EPA policy. Abstract and Poster for 2019 Society of Toxicology annual meeting in March 2019
Bioinformatic Integration of in vivo Data and Literature-based Gene Associations for Prioritization of Adverse Outcome Pathway Development
Adverse outcome pathways (AOPs) describe a sequence of events, beginning with a molecular initiating event (MIE), proceeding via key events (KEs), and culminating in an adverse outcome (AO). A challenge for use of AOPs in a safety evaluation context has been identification of MIEs and KEs relevant for AOs observed in regulatory toxicity studies. In this work, we implemented a bioinformatic approach that leverages mechanistic information in the literature and the AOs measured in regulatory toxicity studies to prioritize putative MIEs and/or early KEs for AOP development relevant to chemical safety evaluation. The US Environmental Protection Agency Toxicity Reference Database (ToxRefDB, v2.0) contains effect information for >1000 chemicals curated from >5000 studies or summaries from sources including data evaluation records from the US EPA Office of Pesticide Programs, the National Toxicology Program (NTP), peer-reviewed literature, and pharmaceutical preclinical studies. To increase ToxRefDB interoperability, endpoint and effect information were cross-referenced with codes from the United Medical Language System, which enabled mapping of in vivo pathological effects from ToxRefDB to PubMed (via Medical Subject Headings or MeSH). This enabled linkage to any resource that is also connected to PubMed or indexed with MeSH. A publicly available bioinformatic tool, the Entity-MeSH Co-occurrence Network (EMCON), uses multiple data sources and a measure of mutual information to identify genes most related to a MeSH term. Using EMCON, gene sets were generated for endpoints of toxicological relevance in ToxRefDB linking putative KEs and/or MIEs. The Comparative Toxicogenomics Database was used to further filter important associations. As a proof of concept, thyroid-related effects and their highly associated genes were examined, and demonstrated relevant MIEs and early KEs for AOPs to describe thyroid-related AOs. The ToxRefDB to gene mapping for thyroid resulted in >50 unique gene to chemical relationships. Integrated use of EMCON and ToxRefDB data provides a basis for rapid and robust putative AOP development, as well as a novel means to generate mechanistic hypotheses for specific chemicals. This abstract does not necessarily reflect U.S. EPA policy. Abstract and Poster for 2019 Society of Toxicology annual meeting in March 2019
A Web-Hosted R Workflow to Simplify and Automate the Analysis of 16S NGS Data
Next-Generation Sequencing (NGS) produces large data sets that include tens-of-thousands of sequence reads per sample. For analysis of bacterial diversity, 16S NGS sequences are typically analyzed in a workflow that containing best-of-breed bioinformatics packages that may leverage multiple programming languages (e.g., Python, R, Java, etc.). The process totransform raw NGS data to usable operational taxonomic units (OTUs) can be tedious due tothe number of quality control (QC) steps used in QIIME and other software packages forsample processing. Therefore, the purpose of this work was to simplify the analysis of 16SNGS data from a large number of samples by integrating QC, demultiplexing, and QIIME(Quantitative Insights Into Microbial Ecology) analysis in an accessible R project. User command line operations for each of the pipeline steps were automated into a workflow. In addition, the R server allows multi-user access to the automated pipeline via separate useraccounts while providing access to the same large set of underlying data. We demonstratethe applicability of this pipeline automation using 16S NGS data from approximately 100 stormwater runoff samples collected in a mixed-land use watershed in northeast Georgia. OTU tables were generated for each sample and the relative taxonomic abundances were compared for different periods over storm hydrographs to determine how the microbial ecology of a stream changes with rise and fall of stream stage. Our approach simplifies the pipeline analysis of multiple 16S NGS samples by automating multiple preprocessing, QC, analysis and post-processing command line steps that are called by a sequence of R scripts. Presented at ASM 2015 Rapid NGS Bioinformatic Pipelines for Enhanced Molecular Epidemiologic Investigation of Pathogens
DEVELOPING COMPUTATIONAL TOOLS NECESSARY FOR APPLYING TOXICOGENOMICS TO RISK ASSESSMENT AND REGULATORY DECISION MAKING.
GENOMICS, PROTEOMICS & METABOLOMICS CAN PROVIDE USEFUL WEIGHT-OF-EVIDENCE DATA ALONG THE SOURCE-TO-OUTCOME CONTINUUM, WHEN APPROPRIATE BIOINFORMATIC AND COMPUTATIONAL METHODS ARE APPLIED TOWARDS INTEGRATING MOLECULAR, CHEMICAL AND TOXICOGICAL INFORMATION. GENOMICS, PROTEOMICS & METABOLOMICS CAN PROVIDE USEFUL WEIGHT-OF-EVIDENCE DATA ALONG THE SOURCE-TO-OUTCOME CONTINUUM, WHEN APPROPRIATE BIOINFORMATIC AND COMPUTATIONAL METHODS ARE APPLIED TOWARDS INTEGRATING MOLECULAR, CHEMICAL AND TOXICOGICAL INFORMATION.
The Human Toxome project, funded as an NIH Transformative Research grant 2011–‐ 2016, is focused on developing the concepts and the means for deducing, validating, and sharing molecular Pathways of Toxicity (PoT). Using the test case of estrogenic endocrine disruption, the responses of MCF–‐7 human breast cancer cells are being phenotyped by transcriptomics and mass–‐spectroscopy–‐based metabolomics. The bioinformatics tools for PoT deduction represent a core deliverable. A number of challenges for quality and standardization of cell systems, omics technologies, and bioinformatics are being addressed. In parallel, concepts for annotation, validation, and sharing of PoT information, as well as their link to adverse outcomes, are being developed. A reasonably comprehensive public database of PoT, the Human Toxome Knowledge–‐base, could become a point of reference for toxicological research and regulatory tests strategies. A reasonably comprehensive public database of PoT, the Human Toxome Knowledge–‐base, could become a point of reference for toxicological research and regulatory tests strategies.
High-Resolution Metabolomics for Environmental Chemical Surveillance and Bioeffect Monitoring
High-Resolution Metabolomics for Environmental Chemical Surveillance and Bioeffect Monitoring (Presented by: Dean Jones, PhD, Department of Medicine, Emory University) (2/28/2013)
Identification of Absorption, Distribution, Metabolism, and Excretion (ADME) Genes Relevant to Steatosis Using a Gene Expression Approach
Absorption, distribution, metabolism, and excretion (ADME) impact chemical concentration and activation of molecular initiating events of Adverse Outcome Pathways (AOPs) in cellular, tissue, and organ level targets. In order to better describe ADME parameters and how they modulate potential hazards posed by chemical exposure, our goal is to investigate the relationship between AOPs and ADME related genes and functional information. Given the scope of this task, we began using hepatic steatosis as a case study. To identify ADME genes related to steatosis, we used the publicly available toxicogenomics database, Open TG-GATEsTM. This database contains standardized rodent chemical exposure data from 170 chemicals (mostly drugs), along with differential gene expression data and corresponding associated pathological changes. We examined the chemical exposure microarray data set gathered from 9 chemical exposure treatments resulting in pathologically confirmed (minimal, moderate and severe) incidences of hepatic steatosis. From this differential gene expression data set, we utilized differential expression analyses to identify gene changes resulting from the chemical exposures leading to hepatic steatosis. We then selected differentially expressed genes (DEGs) related to ADME by filtering all genes based on their ADME functional identities. These DEGs include enzymes such as cytochrome p450, UDP glucuronosyltransferase, flavin-containing monooxygenase and transporter genes such as solute carriers and ATP-binding cassette transporter families. The up and downregulated genes were identified across these treatments. Total of 61 genes were upregulated and 68 genes were down regulated in all treatments. Meanwhile, 25 genes were both up regulated and downregulated across all the treatments. This work highlights the application of bioinformatics in linking AOPs with gene modulations specifically in relationships to ADME and exposures to chemicals. This abstract does not necessarily reflect U.S. EPA policy. This work highlights the application of bioinformatics tools to identify genes that are modulated by adverse outcomes. Specifically, we delineate a method to identify genes that are related to ADME and can impact target tissue dose in response to chemical exposures. The computational method outlined in this work is applicable to any adverse outcome pathway, and provide a linkage between chemical exposure, target tissue dose, and adverse outcomes. Application of this method will allow for the rapid screening of chemicals for their impact on ADME-related genes using available gene data bases in literature.
EPA’s Computational Toxicology Communities of Practice is composed of hundreds of stakeholders from over 50 public and private sector organizations (ranging from EPA, other federal agencies, industry, academic institutions, professional societies, nongovernmental organizations, environmental non-profit groups, state environmental agencies and more) who have an interest in using advances in computational toxicology and exposure science to evaluate the safety of chemicals.
The Communities of Practice is open to the public. Monthly webinars are held at EPA’s RTP campus, on the fourth Thursday of the month (occasionally rescheduled in November and December to accommodate holiday schedules), from 11am-Noon EST/EDT. Remote participation is available. For more information or to be added to the meeting email list, contact: Monica Linnenbrink (linnenbrink.monica@epa.gov).
Dr. Antony Williams, Chemist in EPA’s National Center for Computational Toxicology and Dr. Katie Paul-Friedman, Toxicologist in EPA’s National Center for Computational Toxicology
Colleen Elonen, Translational Toxicology Branch, and Dr. Jennifer Olker, Systems Toxicology Branch, in the Mid-Continent Ecology Division of EPA’s National Health & Environmental Effects Research Laboratory (NHEERL)
2018/05/24
Investigating Chemical-Microbiota Interactions in Zebrafish (VideoEXIT)
Tamara Tal, Biologist in the Systems Biology Branch, Integrated Systems Toxicology Division, EPA’s National Health & Environmental Effects Research Laboratory (NHEERL)
Katherine Phillips, Research Chemist, Human Exposure and Dose Modeling Branch, Computational Exposure Division, EPA’s National Exposure Research Laboratory (NERHL)
2012pharmaceutical
Amandeep Kaur
Aashir Awan, Phd
Abhisar Anand
Adina Hazan
Alan F. Kaul, PharmD., MS, MBA, FCCP
alexcrystal6
anamikasarkar
apreconasia
aptubman
Arav Gandhi
aviralvatsa
David Orchard-Webb, PhD
danutdaagmailcom
Demet Sag, Ph.D., CRA, GCP
Dror Nir
dsmolyar
Ethan Coomber
evelinacohn
Gail S Thornton
Irina Robu
jayzmit48
jdpmd
jshertok
kellyperlman
Ed Kislauskis
larryhbern
Madison Davis
marzankhan
megbaker58
ofermar2020
Dr. Pati
pkandala
ritusaxena
Rick Mandahl
sjwilliamspa
Srinivas Sriram
stuartlpbi
Dr. Sudipta Saha
tildabarliya
zraviv06
zs22