Predicting human genetic interactions from cancer genome evolution

Synthetic lethal connectivity and graph transformer improve synthetic lethality prediction

Synthetic lethality (SL) has shown great promise for the discovery of novel targets in cancer. CRISPR double-knockout (CDKO) technologies can only screen several hundred genes and their combinations, but not genome-wide. Therefore, good SL prediction models are highly needed for genes and gene pairs selection in CDKO experiments. However, lack of scalable SL properties prevents generalizability of SL interactions to out-of-sample data, thereby hindering modeling efforts. In this paper, we recognize that SL connectivity is a scalable and generalizable SL property. We develop a novel two-step multilayer encoder for individual sample-specific SL prediction model (MLEC-iSL), which predicts SL connectivity first and SL interactions subsequently. MLEC-iSL has three encoders, namely, gene, graph, and transformer encoders. MLEC-iSL achieves high SL prediction performance in K562 (AUPR, 0.73; AUC, 0.72) and Jurkat (AUPR, 0.73; AUC, 0.71) cells, while no existing methods exceed 0.62 AUPR and AUC. The prediction performance of MLEC-iSL is validated in a CDKO experiment in 22Rv1 cells, yielding a 46.8% SL rate among 987 selected gene pairs. The screen also reveals SL dependency between apoptosis and mitosis cell death pathways.

CSSLdb: Discovery of cancer-specific synthetic lethal interactions based on machine learning and statistic inference

Synthetic lethality (SL) occurs when the inactivation of two genes results in cell death while the inactivation of either gene alone is non-lethal. SL-based therapy has become a promising anti-cancer treatment option with the increasing researches and applications in clinical practice, while the specific therapeutic opportunities for various cancers have not yet been comprehensively investigated. Herein, we described a computational approach based on machine learning and statistical inference to discover the cancer-specific synthetic lethal interactions. First, Random Forest and One-Class SVM were used to perform cancer unbiased prediction of synthetic lethality. Then, two strategies, including mutual exclusivity and differential expression, were used to screen cancer-specific synthetic lethal interactions, resulting in 14,582 SL gene pairs in 33 cancer types. Finally, we developed a freely available database of CSSLdb (Cancer Specific Synthetic Lethality Database, http://www.tmliang.cn/CSSL/) to present cancer-specific synthetic lethal genetic interactions, which would enrich the relevant research and contribute to underlying therapy strategies based on synthetic lethality.

ELISL: early-late integrated synthetic lethality prediction in cancer

MOTIVATION

Anti-cancer therapies based on synthetic lethality (SL) exploit tumour vulnerabilities for treatment with reduced side effects, by targeting a gene that is jointly essential with another whose function is lost. Computational prediction is key to expedite SL screening, yet existing methods are vulnerable to prevalent selection bias in SL data and reliant on cancer or tissue type-specific omics, which can be scarce. Notably, sequence similarity remains underexplored as a proxy for related gene function and joint essentiality.

RESULTS

We propose ELISL, Early-Late Integrated SL prediction with forest ensembles, using context-free protein sequence embeddings and context-specific omics from cell lines and tissue. Across eight cancer types, ELISL showed superior robustness to selection bias and recovery of known SL genes, as well as promising cross-cancer predictions. Co-occurring mutations in a BRCA gene and ELISL-predicted pairs from the HH, FGF, WNT, or NEIL gene families were associated with longer patient survival times, revealing therapeutic potential.

AVAILABILITY AND IMPLEMENTATION

Data: 10.6084/m9.figshare.23607558 & Code: github.com/joanagoncalveslab/ELISL.

Multi-view graph convolutional network for cancer cell-specific synthetic lethality prediction

Synthetic lethal (SL) genetic interactions have been regarded as a promising focus for investigating potential targeted therapeutics to tackle cancer. However, the costly investment of time and labor associated with wet-lab experimental screenings to discover potential SL relationships motivates the development of computational methods. Although graph neural network (GNN) models have performed well in the prediction of SL gene pairs, existing GNN-based models are not designed for predicting cancer cell-specific SL interactions that are more relevant to experimental validation in vitro. Besides, neither have existing methods fully utilized diverse graph representations of biological features to improve prediction performance. In this work, we propose MVGCN-iSL, a novel multi-view graph convolutional network (GCN) model to predict cancer cell-specific SL gene pairs, by incorporating five biological graph features and multi-omics data. Max pooling operation is applied to integrate five graph-specific representations obtained from GCN models. Afterwards, a deep neural network (DNN) model serves as the prediction module to predict the SL interactions in individual cancer cells (iSL). Extensive experiments have validated the model's successful integration of the multiple graph features and state-of-the-art performance in the prediction of potential SL gene pairs as well as generalization ability to novel genes.

Synthetic lethal gene pairs: Experimental approaches and predictive models

Synthetic lethality (SL) refers to a genetic interaction in which the simultaneous perturbation of two genes leads to cell or organism death, whereas viability is maintained when only one of the pair is altered. The experimental exploration of these pairs and predictive modeling in computational biology contribute to our understanding of cancer biology and the development of cancer therapies. We extensively reviewed experimental technologies, public data sources, and predictive models in the study of synthetic lethal gene pairs and herein detail biological assumptions, experimental data, statistical models, and computational schemes of various predictive models, speculate regarding their influence on individual sample- and population-based synthetic lethal interactions, discuss the pros and cons of existing SL data and models, and highlight potential research directions in SL discovery.

Overcoming selection bias in synthetic lethality prediction

MOTIVATION

Synthetic lethality (SL) between two genes occurs when simultaneous loss of function leads to cell death. This holds great promise for developing anti-cancer therapeutics that target synthetic lethal pairs of endogenously disrupted genes. Identifying novel SL relationships through exhaustive experimental screens is challenging, due to the vast number of candidate pairs. Computational SL prediction is therefore sought to identify promising SL gene pairs for further experimentation. However, current SL prediction methods lack consideration for generalizability in the presence of selection bias in SL data.

RESULTS

We show that SL data exhibit considerable gene selection bias. Our experiments designed to assess the robustness of SL prediction reveal that models driven by the topology of known SL interactions (e.g. graph, matrix factorization) are especially sensitive to selection bias. We introduce selection bias-resilient synthetic lethality (SBSL) prediction using regularized logistic regression or random forests. Each gene pair is described by 27 molecular features derived from cancer cell line, cancer patient tissue and healthy donor tissue samples. SBSL models are built and tested using approximately 8000 experimentally derived SL pairs across breast, colon, lung and ovarian cancers. Compared to other SL prediction methods, SBSL showed higher predictive performance, better generalizability and robustness to selection bias. Gene dependency, quantifying the essentiality of a gene for cell survival, contributed most to SBSL predictions. Random forests were superior to linear models in the absence of dependency features, highlighting the relevance of mutual exclusivity of somatic mutations, co-expression in healthy tissue and differential expression in tumour samples.

AVAILABILITY AND IMPLEMENTATION

https://github.com/joanagoncalveslab/sbsl.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Combination of Niraparib, Cisplatin and Twist Knockdown in Cisplatin-Resistant Ovarian Cancer Cells Potentially Enhances Synthetic Lethality through ER-Stress Mediated Mitochondrial Apoptosis Pathway

Poly (ADP-ribose) polymerase 1 inhibitors (PARPi) are used to treat recurrent ovarian cancer (OC) patients due to greater survival benefits and minimal side effects, especially in those patients with complete or partial response to platinum-based chemotherapy. However, acquired resistance of platinum-based chemotherapy leads to the limited efficacy of PARPi monotherapy in most patients. Twist is recognized as a possible oncogene and contributes to acquired cisplatin resistance in OC cells. In this study, we show how Twist knockdown cisplatin-resistant (CisR) OC cells blocked DNA damage response (DDR) to sensitize these cells to a concurrent treatment of cisplatin as a platinum-based chemotherapy agent and niraparib as a PARPi on in vitro two-dimensional (2D) and three-dimensional (3D) cell culture. To investigate the lethality of PARPi and cisplatin on Twist knockdown CisR OC cells, two CisR cell lines (OV90 and SKOV3) were established using step-wise dose escalation method. In addition, in vitro 3D spheroidal cell model was generated using modified hanging drop and hydrogel scaffolds techniques on poly-2-hydroxylethly methacrylate (poly-HEMA) coated plates. Twist expression was strongly correlated with the expression of DDR proteins, PARP1 and XRCC1 and overexpression of both proteins was associated with cisplatin resistance in OC cells. Moreover, combination of cisplatin (Cis) and niraparib (Nira) produced lethality on Twist-knockdown CisR OC cells, according to combination index (CI). We found that Cis alone, Nira alone, or a combination of Cis+Nira therapy increased cell death by suppressing DDR proteins in 2D monolayer cell culture. Notably, the combination of Nira and Cis was considerably effective against 3D-cultures of Twist knockdown CisR OC cells in which Endoplasmic reticulum (ER) stress is upregulated, leading to initiation of mitochondrial-mediated cell death. In addition, immunohistochemically, Cis alone, Nira alone or Cis+Nira showed lower ki-67 (cell proliferative marker) expression and higher cleaved caspase-3 (apoptotic marker) immuno-reactivity. Hence, lethality of PARPi with the combination of Cis on Twist knockdown CisR OC cells may provide an effective way to expand the therapeutic potential to overcome platinum-based chemotherapy resistance and PARPi cross resistance in OC.

Combinatorial CRISPR screen identifies fitness effects of gene paralogues

Genetic redundancy has evolved as a way for human cells to survive the loss of genes that are single copy and essential in other organisms, but also allows tumours to survive despite having highly rearranged genomes. In this study we CRISPR screen 1191 gene pairs, including paralogues and known and predicted synthetic lethal interactions to identify 105 gene combinations whose co-disruption results in a loss of cellular fitness. 27 pairs influence fitness across multiple cell lines including the paralogues FAM50A/FAM50B, two genes of unknown function. Silencing of FAM50B occurs across a range of tumour types and in this context disruption of FAM50A reduces cellular fitness whilst promoting micronucleus formation and extensive perturbation of transcriptional programmes. Our studies reveal the fitness effects of FAM50A/FAM50B in cancer cells.

Predicting synthetic lethal interactions using heterogeneous data sources

MOTIVATION

A synthetic lethal (SL) interaction is a relationship between two functional entities where the loss of either one of the entities is viable but the loss of both entities is lethal to the cell. Such pairs can be used as drug targets in targeted anticancer therapies, and so, many methods have been developed to identify potential candidate SL pairs. However, these methods use only a subset of available data from multiple platforms, at genomic, epigenomic and transcriptomic levels; and hence are limited in their ability to learn from complex associations in heterogeneous data sources.

RESULTS

In this article, we develop techniques that can seamlessly integrate multiple heterogeneous data sources to predict SL interactions. Our approach obtains latent representations by collective matrix factorization-based techniques, which in turn are used for prediction through matrix completion. Our experiments, on a variety of biological datasets, illustrate the efficacy and versatility of our approach, that outperforms state-of-the-art methods for predicting SL interactions and can be used with heterogeneous data sources with minimal feature engineering.

AVAILABILITY AND IMPLEMENTATION

Software available at https://github.com/lianyh.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

G2G: A web-server for the prediction of human synthetic lethal interactions

Genetic interactions (GIs) are fundamental to our understanding of biological processes in the cell. While GIs have been systematically mapped in yeast, there is scarce information about them in humans. Recently, we have suggested a state-of-the-art hierarchical method that leverages gene ontology information for predicting GIs in yeast. Here, we adapt this method and apply it for the first time to predict GIs in human. We introduce a web service called G2G for this task that is available at http://bnet.cs.tau.ac.il/g2g/.

Identifying Epistasis in Cancer Genomes: A Delicate Affair

Recent studies of the tumor genome seek to identify cancer pathways as groups of genes in which mutations are epistatic with one another or, specifically, "mutually exclusive." Here, we show that most mutations are mutually exclusive not due to pathway structure but to interactions with disease subtype and tumor mutation load. In particular, many cancer driver genes are mutated preferentially in tumors with few mutations overall, causing mutations in these cancer genes to appear mutually exclusive with numerous others. Researchers should view current epistasis maps with caution until we better understand the multiple cause-and-effect relationships among factors such as tumor subtype, positive selection for mutations, and gross tumor characteristics including mutational signatures and load.

Pan-Cancer Analysis of Potential Synthetic Lethal Drug Targets Specific to Alterations in DNA Damage Response

Alterations in DNA damage response (DDR) is one of the several hallmarks of cancer. Genomic instability resulting from a disrupted DDR mechanism is known to contribute to cancer progression, and are subjected to radiation, cytotoxic, or more recently targeted therapies with limited success. Synthetic lethality (SL), which is a condition where simultaneous loss-of-function of the genes from complementary pathways result in loss of viability of cancer cells have been exploited to treat malignancies resulting from defects in certain DDR pathways. Albeit being a promising therapeutic strategy, number of SL based drugs currently in clinical trial is limited. In this work we performed a comprehensive pan-cancer analysis of alterations in 10 DDR pathways with different components of DNA repair. Using unsupervised clustering of single sample enrichment of these pathways in 7,272 tumor samples from 17 tumor types from TCGA, we identified three prominent clusters, each associated with specific DDR mechanisms. Somatic mutations in key DDR genes were found to be dominant in each of these three clusters with distinct DDR component. Using a machine-learning based algorithm we predicted SL partners specific to somatic mutations in key genes representing each of the three DDR clusters and identified potential druggable targets. We explored the potential FDA-approved drugs for targeting the predicted SL genes and tested the sensitivity using the drug screening data in cell lines with mutation in the primary DDR genes. We have shown clinical relevance, for selected targetable SL interactions using Kaplan-Meier analysis in terms of improved disease-free survival. Thus, our computational framework provides a basis for clinically relevant and actionable SL based drug targets specific to alterations in DDR pathways.

Harnessing synthetic lethality to predict the response to cancer treatment

While synthetic lethality (SL) holds promise in developing effective cancer therapies, SL candidates found via experimental screens often have limited translational value. Here we present a data-driven approach, ISLE (identification of clinically relevant synthetic lethality), that mines TCGA cohort to identify the most likely clinically relevant SL interactions (cSLi) from a given candidate set of lab-screened SLi. We first validate ISLE via a benchmark of large-scale drug response screens and by predicting drug efficacy in mouse xenograft models. We then experimentally test a select set of predicted cSLi via new screening experiments, validating their predicted context-specific sensitivity in hypoxic vs normoxic conditions and demonstrating cSLi's utility in predicting synergistic drug combinations. We show that cSLi can successfully predict patients' drug treatment response and provide patient stratification signatures. ISLE thus complements existing actionable mutation-based methods for precision cancer therapy, offering an opportunity to expand its scope to the whole genome.

Data integration and predictive modeling methods for multi-omics datasets

We provide an overview of opportunities and challenges in multi-omics predictive analytics with particular emphasis on data integration and machine learning methods.

Synthetic lethality: emerging targets and opportunities in melanoma

Great progress has been made in the treatment of melanoma through use of targeted therapies and immunotherapy. One approach that has not been fully explored is synthetic lethality, which exploits somatically acquired changes, usually driver mutations, to specifically kill tumour cells. We outline the various approaches that may be applied to identify synthetic lethal interactions and define how these interactions may drive drug discovery efforts.

Multi-omic measurement of mutually exclusive loss-of-function enriches for candidate synthetic lethal gene pairs

Background

Identification of synthetic lethal interactions in cancer cells could offer promising new therapeutic targets. Large-scale functional genomic screening presents an opportunity to test large numbers of cancer synthetic lethal hypotheses. Methods enriching for candidate synthetic lethal targets in molecularly defined cancer cell lines can steer effective design of screening efforts. Loss of one partner of a synthetic lethal gene pair creates a dependency on the other, thus synthetic lethal gene pairs should never show simultaneous loss-of-function. We have developed a computational approach to mine large multi-omic cancer data sets and identify gene pairs with mutually exclusive loss-of-function. Since loss-of-function may not always be genetic, we look for deleterious mutations, gene deletion and/or loss of mRNA expression by bimodality defined with a novel algorithm BiSEp.

Results

Applying this toolkit to both tumour cell line and patient data, we achieve statistically significant enrichment for experimentally validated tumour suppressor genes and synthetic lethal gene pairings. Notably non-reliance on genetic loss reveals a number of known synthetic lethal relationships otherwise missed, resulting in marked improvement over genetic-only predictions. We go on to establish biological rationale surrounding a number of novel candidate synthetic lethal gene pairs with demonstrated dependencies in published cancer cell line shRNA screens.

Conclusions

This work introduces a multi-omic approach to define gene loss-of-function, and enrich for candidate synthetic lethal gene pairs in cell lines testable through functional screens. In doing so, we offer an additional resource to generate new cancer drug target and combination hypotheses. Algorithms discussed are freely available in the BiSEp CRAN package at http://cran.r-project.org/web/packages/BiSEp/index.html.

Electronic supplementary material

The online version of this article (doi:10.1186/s12864-016-2375-1) contains supplementary material, which is available to authorized users.

Oncogenomic portals for the visualization and analysis of genome-wide cancer data

Somatically acquired genomic alterations that drive oncogenic cellular processes are of great scientific and clinical interest. Since the initiation of large-scale cancer genomic projects (e.g., the Cancer Genome Project, The Cancer Genome Atlas, and the International Cancer Genome Consortium cancer genome projects), a number of web-based portals have been created to facilitate access to multidimensional oncogenomic data and assist with the interpretation of the data. The portals provide the visualization of small-size mutations, copy number variations, methylation, and gene/protein expression data that can be correlated with the available clinical, epidemiological, and molecular features. Additionally, the portals enable to analyze the gathered data with the use of various user-friendly statistical tools. Herein, we present a highly illustrated review of seven portals, i.e., Tumorscape, UCSC Cancer Genomics Browser, ICGC Data Portal, COSMIC, cBioPortal, IntOGen, and BioProfiling.de. All of the selected portals are user-friendly and can be exploited by scientists from different cancer-associated fields, including those without bioinformatics background. It is expected that the use of the portals will contribute to a better understanding of cancer molecular etiology and will ultimately accelerate the translation of genomic knowledge into clinical practice.

Prediction of Genetic Interactions Using Machine Learning and Network Properties

A genetic interaction (GI) is a type of interaction where the effect of one gene is modified by the effect of one or several other genes. These interactions are important for delineating functional relationships among genes and their corresponding proteins, as well as elucidating complex biological processes and diseases. An important type of GI - synthetic sickness or synthetic lethality - involves two or more genes, where the loss of either gene alone has little impact on cell viability, but the combined loss of all genes leads to a severe decrease in fitness (sickness) or cell death (lethality). The identification of GIs is an important problem for it can help delineate pathways, protein complexes, and regulatory dependencies. Synthetic lethal interactions have important clinical and biological significance, such as providing therapeutically exploitable weaknesses in tumors. While near systematic high-content screening for GIs is possible in single cell organisms such as yeast, the systematic discovery of GIs is extremely difficult in mammalian cells. Therefore, there is a great need for computational approaches to reliably predict GIs, including synthetic lethal interactions, in these organisms. Here, we review the state-of-the-art approaches, strategies, and rigorous evaluation methods for learning and predicting GIs, both under general (healthy/standard laboratory) conditions and under specific contexts, such as diseases.

Synthetic dosage lethality in the human metabolic network is highly predictive of tumor growth and cancer patient survival

Synthetic dosage lethality (SDL) denotes a genetic interaction between two genes whereby the underexpression of gene A combined with the overexpression of gene B is lethal. SDLs offer a promising way to kill cancer cells by inhibiting the activity of SDL partners of activated oncogenes in tumors, which are often difficult to target directly. As experimental genome-wide SDL screens are still scarce, here we introduce a network-level computational modeling framework that quantitatively predicts human SDLs in metabolism. For each enzyme pair (A, B) we systematically knock out the flux through A combined with a stepwise flux increase through B and search for pairs that reduce cellular growth more than when either enzyme is perturbed individually. The predictive signal of the emerging network of 12,000 SDLs is demonstrated in five different ways. (i) It can be successfully used to predict gene essentiality in shRNA cancer cell line screens. Moving to clinical tumors, we show that (ii) SDLs are significantly underrepresented in tumors. Furthermore, breast cancer tumors with SDLs active (iii) have smaller sizes and (iv) result in increased patient survival, indicating that activation of SDLs increases cancer vulnerability. Finally, (v) patient survival improves when multiple SDLs are present, pointing to a cumulative effect. This study lays the basis for quantitative identification of cancer SDLs in a model-based mechanistic manner. The approach presented can be used to identify SDLs in species and cell types in which "omics" data necessary for data-driven identification are missing.