Avinash Das

Bayesian integration of genetics and epigenetics detects causal regulatory SNPs underlying expression variability

The standard expression quantitative trait loci (eQTL) detects polymorphisms associated with gene expression without revealing causality. We introduce a coupled Bayesian regression approach—eQTeL, which leverages epigenetic data to estimate regulatory and gene interaction potential, and identifies combination of regulatory single-nucleotide polymorphisms (SNPs) that explain the gene expression variance. On human heart data, eQTeL not only explains a significantly greater proportion of expression variance but also predicts gene expression more accurately than other methods. Based on realistic simulated data, we demonstrate that eQTeL accurately detects causal regulatory SNPs, including those with small effect sizes. Using various functional data, we show that SNPs detected by eQTeL are enriched for allele-specific protein binding and histone modifications, which potentially disrupt binding of core cardiac transcription factors and are spatially proximal to their target. eQTeL SNPs capture a substantial proportion of genetic determinants of expression variance and we estimate that 58% of these SNPs are putatively causal.

Distal CpG islands can serve as alternative promoters to transcribe genes with silenced proximal promoters

DNA methylation at the promoter of a gene is presumed to render it silent, yet a sizable fraction of genes with methylated proximal promoters exhibit elevated expression. Here, we show, through extensive analysis of the methylome and transcriptome in 34 tissues, that in many such cases, transcription is initiated by a distal upstream CpG island (CGI) located several kilobases away that functions as an alternative promoter. Specifically, such genes are expressed precisely when the neighboring CGI is unmethylated but remain silenced otherwise. Based on CAGE and Pol II localization data, we found strong evidence of transcription initiation at the upstream CGI and a lack thereof at the methylated proximal promoter itself. Consistent with their alternative promoter activity, CGI-initiated transcripts are associated with signals of stable elongation and splicing that extend into the gene body, as evidenced by tissue-specific RNA-seq and other DNA-encoded splice signals. Furthermore, based on both inter- and intra-species analyses, such CGIs were found to be under greater purifying selection relative to CGIs upstream of silenced genes. Overall, our study describes a hitherto unreported conserved mechanism of transcription of genes with methylated proximal promoters in a tissue-specific fashion. Importantly, this phenomenon explains the aberrant expression patterns of some cancer driver genes, potentially due to aberrant hypomethylation of distal CGIs, despite methylation at proximal promoters.

Prediction and Subtyping of Hypertension from Pan-Tissue Transcriptomic and Genetic Analyses

Hypertension (HT) is a complex systemic disease involving transcriptional changes in multiple organs. Here we systematically investigate the pan-tissue transcriptional and genetic landscape of HT spanning dozens of tissues in hundreds of individuals. We find that in several tissues, previously identified HT-linked genes are dysregulated and the gene expression profile is predictive of HT. Importantly, many expression quantitative trait loci (eQTL) SNPs associated with the population variance of the dysregulated genes are linked with blood pressure in an independent genome-wide association study, suggesting that the functional effect of HT-associated SNPs may be mediated through tissue-specific transcriptional dysregulation. Analyses of pan-tissue transcriptional dysregulation profile, as well as eQTL SNPs underlying the dysregulated genes, reveals substantial heterogeneity among the HT patients, revealing two broad groupings – a Diffused group where several tissues exhibit HT-associated molecular alterations and a Localized group where such alterations are localized to very few tissues. These two patient subgroups differ in several clinical phenotypes including respiratory, cerebrovascular, diabetes, and heart disease. These findings suggest that the Diffused and Localized subgroups may be driven by different molecular mechanisms and have different genetic underpinning.

Putative functional genes in idiopathic dilated cardiomyopathy

Idiopathic dilated cardiomyopathy (DCM) is a complex disorder with a genetic and an environmental component involving multiple genes, many of which are yet to be discovered. We integrate genetic, epigenetic, transcriptomic, phenotypic, and evolutionary features into a method – Hridaya, to infer putative functional genes underlying DCM in a genome-wide fashion, using 213 human heart genomes and transcriptomes. Many genes identified by Hridaya are experimentally shown to cause cardiac complications. We validate the top predicted genes, via five different genome-wide analyses: First, the predicted genes are associated with cardiovascular functions. Second, their knockdowns in mice induce cardiac abnormalities. Third, their inhibition by drugs cause cardiac side effects in human. Fourth, they tend to have differential exon usage between DCM and normal samples. Fifth, analyzing 213 individual genotypes, we show that regulatory polymorphisms of the predicted genes are associated with elevated risk of cardiomyopathy. The stratification of DCM patients based on cardiac expression of the functional genes reveals two subgroups differing in key cardiac phenotypes. Integrating predicted functional genes with cardiomyocyte drug treatment experiments reveals novel potential drug targets. We provide a list of investigational drugs that target the newly identified functional genes that may lead to cardiac side effects.

Phenotype-dependent coexpression gene clusters: application to normal and premature ageing

Hutchinson Gilford progeria syndrome (HGPS) is a rare genetic disease with symptoms of aging at a very early age. Its molecular basis is not entirely clear, although profound gene expression changes have been reported, and there are some known and other presumed overlaps with normal aging process. Identification of genes with agingor HGPS-associated expression changes is thus an important problem. However, standard regression approaches are currently unsuitable for this task due to limited sample sizes, thus motivating development of alternative approaches. Here, we report a novel iterative multiple regression approach that leverages co-expressed gene clusters to identify gene clusters whose expression co-varies with age and/or HGPS. We have applied our approach to novel RNA-seq profiles in fibroblast cell cultures at three different cellular ages, both from HGPS patients and normal samples. After establishing the robustness of our approach, we perform a comparative investigation of biological processes underlying normal aging and HGPS. Our results recapitulate previously known processes underlying aging as well as suggest numerous unique processes underlying aging and HGPS. The approach could also be useful in detecting phenotype-dependent co-expression gene clusters in other contexts with limited sample sizes.

Beyond synthetic lethality: multiple gene interaction types play a key functional role in cancer

The phenotypic effect of perturbing a gene9s activity depends on the activity state of other genes, reflecting the fundamental notion that genotype to phenotype linkage is mediated by a network of functionally interacting genes. The vast majority of contemporary investigations have focused on just one type of genetic interactions (GI) - synthetic lethality (SL). However, there may be additional types of GIs whose systematic identification may markedly enrich the molecular and functional characterization of cancer. Here, based on a novel data-driven approach, we identify ~72K GIs of 11 new types, shared across cancers. These GIs are highly predictive of patient survival, stratify breast cancer tumors into refined subtypes, and explain differences in patients9 response to drugs and cancer driver genes9 tissue-specificity. These results markedly expand the scope of cancer GIs and lay a strong conceptual and computational basis for future studies of additional types of GIs and for their translational applications.The phenotypic effect of perturbing a gene’s activity depends on the activity level of other genes, reflecting the notion that phenotypes are emergent properties of a network of functionally interacting genes. In the context of cancer, contemporary investigations have primarily focused on just one type of functional genetic interaction (GI) – synthetic lethality (SL). However, there may be additional types of GIs whose systematic identification would enrich the molecular and functional characterization of cancer. Here, we describe a novel data-driven approach called EnGIne, that applied to TCGA data identifies 71,946 GIs spanning 12 distinct types, only a small minority of which are SLs. The detected GIs explain cancer driver genes’ tissue-specificity and differences in patients’ response to drugs, and stratify breast cancer tumors into refined subtypes. These results expand the scope of cancer GIs and lay a conceptual and computational basis for future studies of additional types of GIs and their translational applications. The GI network is accessible online via a web portal [https://amagen.shinyapps.io/cancerapp/].

Migration rather than proliferation transcriptomic signatures are strongly associated with breast cancer patient survival

The efficacy of prospective cancer treatments is routinely estimated by in vitro cell-line proliferation screens. However, it is unclear whether tumor aggressiveness and patient survival are influenced more by the proliferative or the migratory properties of cancer cells. To address this question, we experimentally measured proliferation and migration phenotypes across more than 40 breast cancer cell-lines. Based on the latter, we built and validated individual predictors of breast cancer proliferation and migration levels from the cells’ transcriptomics. We then apply these predictors to estimate the proliferation and migration levels of more than 1000 TCGA breast cancer tumors. Reassuringly, both estimates increase with tumor’s aggressiveness, as qualified by its stage, grade, and subtype. However, predicted tumor migration levels are significantly more strongly associated with patient survival than the proliferation levels. We confirm these finding by conducting siRNA knock-down experiments on the highly migratory MDA-MB-231 cell lines and deriving gene knock-down based proliferation and migration signatures. We show that cytoskeletal drugs might be more beneficial in patients with high predicted migration levels. Taken together, these results testify to the importance of migration levels in determining patient survival.

Comprehensive map of age-associated splicing changes across human tissues and their contributions to age-associated diseases

Alternative splicing contributes to phenotypic diversity at multiple biological scales, and its dysregulation is implicated in both ageing and age-associated diseases in human. Cross-tissue variability in splicing further complicates its links to age-associated phenotypes and elucidating these links requires a comprehensive map of age-associated splicing changes across multiple tissues. Here, we generate such a map by analyzing ~8500 RNA-seq samples across 48 tissues in 544 individuals. Employing a stringent model controlling for multiple confounders, we identify 49,869 tissue-specific age-associated splicing events of 7 distinct types. We find that genome-wide splicing profile is a better predictor of biological age than the gene and transcript expression profiles, and furthermore, age-associated splicing provides additional independent contribution to age-associated complex diseases. We show that the age-associated splicing changes may be explained, in part, by concomitant age-associated changes of the upstream splicing factors. Finally, we show that our splicing-based model of age can successfully predict the relative ages of cells in 8 of the 10 paired longitudinal data as well as in 2 sets of cell passage data. Our study presents the first systematic investigation of age-associated splicing changes across tissues, and further strengthening the links between age-associated splicing and age-associated diseases.

Abstract A023: Cell migration is a stronger predictor of patient survival in breast cancer than cell proliferation

Significance: The efficacy of anticancer drugs has been conventionally estimated in the laboratory by measuring post-treatment in vitro proliferation rates of cancer cell lines. The results of such in vitro measurements, however, rarely translate to human trials, thus posing a significant translational challenge. Additional cellular phenotypes have hence been studied in recent years, examining cell migration and invasion. Understanding how each of these in vitro measured phenotypes contributes to the patient response and survival is hence an important open challenge. Here, mining thousands of breast cancer tumors and cell-line experiments, we explore this relationship and delineate the individual contributions of these phenotypes, in predicting patient survival and response. Methods: Migration and proliferation was measured for 43 different breast cancer cell lines. Integrating these measurements with the cell lines9 transcriptomics, we built gene expression-based predictors of each of these phenotypes in cell lines and in tumors. The predicted phenotypes were then used to study their contribution to patient survival. Results: Analyzing the transcriptomics of these cell lines, we identified specific gene-expression signatures of breast cancer migration and proliferation, that are highly predictive of these phenotypes (using cross validation). Subsequently, we applied these signatures to a collection of more than 2800 breast cancer tumors in the TCGA and METABRIC collection, to predict their proliferation and migration rates. Our analysis shows that both laboratory-measured proliferation and migration signatures are predictive of breast cancer stage, grade, subtypes, and finally, of patient survival. Notably, we find that the predicted migration rates of tumors are stronger predictors of patient survival than their predicted proliferation rates. This finding is further reinforced via analyzing migration and proliferation signatures that we derive from in vitro shRNA knockout experiments. We also find that patients whose tumors have high predicted migration rates specifically benefited from cytoskeletal drug treatments. Finally, we find that the predicted migration rates are associated with response of checkpoint inhibitor in patients. We are further extensively validating this by collecting tumor biopsies from patients post immunotherapy. Conclusions: Taken together, these results testify to the superiority of migration- over proliferation-based transcriptomic signatures in predicting breast cancer tumor phenotypes and patients’ survival. This suggests that in vitro migration measurements of drug response may significantly increase the translational value of cellular phenotypic measurements in predicting drug efficacy in patients. Finally, because tumor migration rates are predictive of cancer immunotherapy, they may provide a viable biomarker for immunotherapy response in patients. Citation Format: Nishanth U. Nair, Avinash Das, Joo Sang Lee, Sridhar Hannenhalli, Sylvia Le Devedec, Bob van de Water, Eytan Ruppin. Cell migration is a stronger predictor of patient survival in breast cancer than cell proliferation [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2017 Oct 26-30; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Ther 2018;17(1 Suppl):Abstract nr A023.

Abstract PR09: Harnessing synthetic lethality to predict clinical outcomes of cancer treatment

Significance: The identification of Synthetic Lethal interactions (SLi) have long been considered a foundation for the advancement of cancer treatment. The rapidly accumulating large-scale patient data now provides a golden opportunity to infer SLi directly from patient samples. Here we present a new data-driven approach termed ISLE for identifying SLi, which is then shown to be predictive of clinical outcomes of cancer treatment in an unsupervised manner, for the first time. Methods: ISLE consists of four inference steps, analyzing tumor, cell line and gene evolutionary data: It first identifies putative SL gene pairs whose co-inactivation is underrepresented in tumors, testifying that they are selected against. Second, it further prioritizes candidate SL pairs whose co-inactivation is associated with better prognosis in patients, testifying that they may hamper tumor progression. Finally, it eliminates false positive SLi using gene essentiality screens (testifying to causal SLi relations) and prioritizing SLi paired genes with similar evolutionary phylogenetic profiles. Results: We applied ISLE to analyze the TCGA tumor collection and generated the first clinically-derived pan-cancer SL-network, composed of SLi common across many cancer types. We validated that these SLi match the known, experimentally identified SLi (AUC=0.87), and show that the SL-network is predictive of patient survival in an independent breast cancer dataset (METABRIC). Based on the predicted SLi, we predicted drug response in a wide variety of in vitro, mouse xenograft and patient data, altogether encompassing >700 single drugs and >5,000 drug combinations in >1,000 cell lines, 375 xenograft models and >5,000 patient samples. Importantly, these predictions were performed in an unsupervised manner, reducing the known risk of over-fitting the data commonly associated with supervised prediction methods. SL-derived predictions are based on computing an SL-score that estimates the efficacy of a given drug in a given tumor based on the latter9s omics data. The SL-score counts the number of inactive SL-partners of a given drug target(s) in the given tumor, reflecting the notion that a drug is likely to be more effective in tumors where many of its targets9 SL-partners are inactive. The predicted SL-scores show significant correlations (R > 0.4) with large-scale in vitro and in vivo drug response screens for the majority of drugs tested. Based on the conjecture that synergism between drugs may be mediated by underlying SLi between their targets, we additionally provide accurate predictions of drug synergism for both in vitro and in vivo drug combination screens (AUC~0.8). Most importantly, we demonstrate for the first time that an SL-network can successfully predict the treatment outcome in cancer patients in multiple large-scale patient datasets including the TCGA, where SLis successfully predict patients9 response for 75% of cancer drugs. Conclusions: ISLE is predictive of the patients9 response for the majority of current cancer drugs. Of paramount importance, the predictions of ISLE are based on SLi between (potentially) all genes in the cancer genome, thus prioritizing treatments for patients whose tumors do not bear specific actionable mutations in cancer driver genes, offering a novel approach to precision-based cancer therapy. The predictive performance of ISLE is likely to further improve with the expected rapid accumulation of additional cancer omics and clinical phenotypic data. Citation Format: Joo Sang Lee, Avinash Das, Livnat Jerby-Arnon, Dikla Atias, Arnaud Amzallag, Cyril H. Benes, Talia Golan, Eytan Ruppin. Harnessing synthetic lethality to predict clinical outcomes of cancer treatment [abstract]. In: Proceedings of the AACR Precision Medicine Series: Opportunities and Challenges of Exploiting Synthetic Lethality in Cancer; Jan 4-7, 2017; San Diego, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2017;16(10 Suppl):Abstract nr PR09.