Connectivity Homology Enables Inter-Species Network Models of Synthetic Lethality

Pathway-driven analysis of synthetic lethal interactions in cancer using perturbation screens

Synthetic lethality offers a promising approach for developing effective therapeutic interventions in cancer when direct targeting of driver genes is impractical. In this study, we comprehensively analyzed large-scale CRISPR, shRNA, and PRISM screens to identify potential synthetic lethal (SL) interactions in pan-cancer and 12 individual cancer types, using a new computational framework that leverages the biological function and signaling pathway information of key driver genes to mitigate the confounding effects of background genetic alterations in different cancer cell lines. This approach has successfully identified several putative SL interactions, including KRAS-MAP3K2 and APC-TCF7L2 in pan cancer, and CCND1-METTL1, TP53-FRS3, SMO-MDM2, and CCNE1-MTOR in liver, blood, skin, and gastric cancers, respectively. In addition, we proposed several FDA-approved cancer-targeted drugs for various cancer types through PRISM drug screens, such as cabazitaxel for VHL-mutated kidney cancer and alectinib for lung cancer with NRAS or KRAS mutations. Leveraging pathway information can enhance the concordance of shRNA and CRISPR screens and provide clinically relevant findings such as the potential efficacy of dasatinib, an inhibitor of SRC, for colorectal cancer patients with mutations in the WNT signaling pathway. These analyses revealed that taking signaling pathway information into account results in the identification of more promising SL interactions.

Synthetic Lethality Screening with Recursive Feature Machines

Synthetic lethality refers to a genetic interaction where the simultaneous perturbation of gene pairs leads to cell death. Synthetically lethal gene pairs (SL pairs) provide a potential avenue for selectively targeting cancer cells based on genetic vulnerabilities. The rise of large-scale gene perturbation screens such as the Cancer Dependency Map (DepMap) offers the opportunity to identify SL pairs automatically using machine learning. We build on a recently developed class of feature learning kernel machines known as Recursive Feature Machines (RFMs) to develop a pipeline for identifying SL pairs based on CRISPR viability data from DepMap. In particular, we first train RFMs to predict viability scores for a given CRISPR gene knockout from cell line embeddings consisting of gene expression and mutation features. After training, RFMs use a statistical operator known as average gradient outer product to provide weights for each feature indicating the importance of each feature in predicting cellular viability. We subsequently apply correlation-based filters to re-weight RFM feature importances and identify those features that are most indicative of low cellular viability. Our resulting pipeline is computationally efficient, taking under 3 minutes for analyzing all 17, 453 knockouts from DepMap for candidate SL pairs. We show that our pipeline more accurately recovers experimentally verified SL pairs than prior approaches. Moreover, our pipeline finds new candidate SL pairs, thereby opening novel avenues for identifying genetic vulnerabilities in cancer.

Discovery of synthetic lethal interactions from large-scale pan-cancer perturbation screens

The development of cancer therapies is limited by the availability of suitable drug targets. Potential candidate drug targets can be identified based on the concept of synthetic lethality (SL), which refers to pairs of genes for which an aberration in either gene alone is non-lethal, but co-occurrence of the aberrations is lethal to the cell. Here, we present SLIdR (Synthetic Lethal Identification in R), a statistical framework for identifying SL pairs from large-scale perturbation screens. SLIdR successfully predicts SL pairs even with small sample sizes while minimizing the number of false positive targets. We apply SLIdR to Project DRIVE data and find both established and potential pan-cancer and cancer type-specific SL pairs consistent with findings from literature and drug response screening data. We experimentally validate two predicted SL interactions (ARID1A-TEAD1 and AXIN1-URI1) in hepatocellular carcinoma, thus corroborating the ability of SLIdR to identify potential drug targets.

NSF4SL: negative-sample-free contrastive learning for ranking synthetic lethal partner genes in human cancers

MOTIVATION

Detecting synthetic lethality (SL) is a promising strategy for identifying anti-cancer drug targets. Targeting SL partners of a primary gene mutated in cancer is selectively lethal to cancer cells. Due to high cost of wet-lab experiments and availability of gold standard SL data, supervised machine learning for SL prediction has been popular. However, most of the methods are based on binary classification and thus limited by the lack of reliable negative data. Contrastive learning can train models without any negative sample and is thus promising for finding novel SLs.

RESULTS

We propose NSF4SL, a negative-sample-free SL prediction model based on a contrastive learning framework. It captures the characteristics of positive SL samples by using two branches of neural networks that interact with each other to learn SL-related gene representations. Moreover, a feature-wise data augmentation strategy is used to mitigate the sparsity of SL data. NSF4SL significantly outperforms all baselines which require negative samples, even in challenging experimental settings. To the best of our knowledge, this is the first time that SL prediction is formulated as a gene ranking problem, which is more practical than the current formulation as binary classification. NSF4SL is the first contrastive learning method for SL prediction and its success points to a new direction of machine-learning methods for identifying novel SLs.

AVAILABILITY AND IMPLEMENTATION

Our source code is available at https://github.com/JieZheng-ShanghaiTech/NSF4SL.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

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.

SANA: cross-species prediction of Gene Ontology GO annotations via topological network alignment

Topological network alignment aims to align two networks node-wise in order to maximize the observed common connection (edge) topology between them. The topological alignment of two protein–protein interaction (PPI) networks should thus expose protein pairs with similar interaction partners allowing, for example, the prediction of common Gene Ontology (GO) terms. Unfortunately, no network alignment algorithm based on topology alone has been able to achieve this aim, though those that include sequence similarity have seen some success. We argue that this failure of topology alone is due to the sparsity and incompleteness of the PPI network data of almost all species, which provides the network topology with a small signal-to-noise ratio that is effectively swamped when sequence information is added to the mix. Here we show that the weak signal can be detected using multiple stochastic samples of “good” topological network alignments, which allows us to observe regions of the two networks that are robustly aligned across multiple samples. The resulting network alignment frequency (NAF) strongly correlates with GO-based Resnik semantic similarity and enables the first successful cross-species predictions of GO terms based on topology-only network alignments. Our best predictions have an AUPR of about 0.4, which is competitive with state-of-the-art algorithms, even when there is no observable sequence similarity and no known homology relationship. While our results provide only a “proof of concept” on existing network data, we hypothesize that predicting GO terms from topology-only network alignments will become increasingly practical as the volume and quality of PPI network data increase.

Computational methods, databases and tools for synthetic lethality prediction

Synthetic lethality (SL) occurs between two genes when the inactivation of either gene alone has no effect on cell survival but the inactivation of both genes results in cell death. SL-based therapy has become one of the most promising targeted cancer therapies in the last decade as PARP inhibitors achieve great success in the clinic. The key point to exploiting SL-based cancer therapy is the identification of robust SL pairs. Although many wet-lab-based methods have been developed to screen SL pairs, known SL pairs are less than 0.1% of all potential pairs due to large number of human gene combinations. Computational prediction methods complement wet-lab-based methods to effectively reduce the search space of SL pairs. In this paper, we review the recent applications of computational methods and commonly used databases for SL prediction. First, we introduce the concept of SL and its screening methods. Second, various SL-related data resources are summarized. Then, computational methods including statistical-based methods, network-based methods, classical machine learning methods and deep learning methods for SL prediction are summarized. In particular, we elaborate on the negative sampling methods applied in these models. Next, representative tools for SL prediction are introduced. Finally, the challenges and future work for SL prediction are discussed.

Comprehensive prediction of robust synthetic lethality between paralog pairs in cancer cell lines

Pairs of paralogs may share common functionality and, hence, display synthetic lethal interactions. As the majority of human genes have an identifiable paralog, exploiting synthetic lethality between paralogs may be a broadly applicable approach for targeting gene loss in cancer. However, only a biased subset of human paralog pairs has been tested for synthetic lethality to date. Here, by analyzing genome-wide CRISPR screens and molecular profiles of over 700 cancer cell lines, we identify features predictive of synthetic lethality between paralogs, including shared protein-protein interactions and evolutionary conservation. We develop a machine-learning classifier based on these features to predict which paralog pairs are most likely to be synthetic lethal and to explain why. We show that our classifier accurately predicts the results of combinatorial CRISPR screens in cancer cell lines and furthermore can distinguish pairs that are synthetic lethal in multiple cell lines from those that are cell-line specific. A record of this paper's transparent peer review process is included in the supplemental information.

KG4SL: knowledge graph neural network for synthetic lethality prediction in human cancers

Motivation

Synthetic lethality (SL) is a promising gold mine for the discovery of anti-cancer drug targets. Wet-lab screening of SL pairs is afflicted with high cost, batch-effect, and off-target problems. Current computational methods for SL prediction include gene knock-out simulation, knowledge-based data mining and machine learning methods. Most of the existing methods tend to assume that SL pairs are independent of each other, without taking into account the shared biological mechanisms underlying the SL pairs. Although several methods have incorporated genomic and proteomic data to aid SL prediction, these methods involve manual feature engineering that heavily relies on domain knowledge.

Results

Here, we propose a novel graph neural network (GNN)-based model, named KG4SL, by incorporating knowledge graph (KG) message-passing into SL prediction. The KG was constructed using 11 kinds of entities including genes, compounds, diseases, biological processes and 24 kinds of relationships that could be pertinent to SL. The integration of KG can help harness the independence issue and circumvent manual feature engineering by conducting message-passing on the KG. Our model outperformed all the state-of-the-art baselines in area under the curve, area under precision-recall curve and F1. Extensive experiments, including the comparison of our model with an unsupervised TransE model, a vanilla graph convolutional network model, and their combination, demonstrated the significant impact of incorporating KG into GNN for SL prediction.

Availability and implementation

: KG4SL is freely available at https://github.com/JieZheng-ShanghaiTech/KG4SL.

Supplementary information

Supplementary data are available at Bioinformatics online.

Graph contextualized attention network for predicting synthetic lethality in human cancers

MOTIVATION

Synthetic Lethality (SL) plays an increasingly critical role in the targeted anticancer therapeutics. In addition, identifying SL interactions can create opportunities to selectively kill cancer cells without harming normal cells. Given the high cost of wet-lab experiments, in silico prediction of SL interactions as an alternative can be a rapid and cost-effective way to guide the experimental screening of candidate SL pairs. Several matrix factorization-based methods have recently been proposed for human SL prediction. However, they are limited in capturing the dependencies of neighbors. In addition, it is also highly challenging to make accurate predictions for new genes without any known SL partners.

RESULTS

In this work, we propose a novel graph contextualized attention network named GCATSL to learn gene representations for SL prediction. First, we leverage different data sources to construct multiple feature graphs for genes, which serve as the feature inputs for our GCATSL method. Second, for each feature graph, we design node-level attention mechanism to effectively capture the importance of local and global neighbors and learn local and global representations for the nodes, respectively. We further exploit multi-layer perceptron (MLP) to aggregate the original features with the local and global representations and then derive the feature-specific representations. Third, to derive the final representations, we design feature-level attention to integrate feature-specific representations by taking the importance of different feature graphs into account. Extensive experimental results on three datasets under different settings demonstrated that our GCATSL model outperforms 14 state-of-the-art methods consistently. In addition, case studies further validated the effectiveness of our proposed model in identifying novel SL pairs.

AVAILABILITY

Python codes and dataset are freely available on GitHub (https://github.com/longyahui/GCATSL) and Zenodo (https://zenodo.org/record/4522679) under the MIT license.

Dual-dropout graph convolutional network for predicting synthetic lethality in human cancers

MOTIVATION

Synthetic lethality (SL) is a promising form of gene interaction for cancer therapy, as it is able to identify specific genes to target at cancer cells without disrupting normal cells. As high-throughput wet-lab settings are often costly and face various challenges, computational approaches have become a practical complement. In particular, predicting SLs can be formulated as a link prediction task on a graph of interacting genes. Although matrix factorization techniques have been widely adopted in link prediction, they focus on mapping genes to latent representations in isolation, without aggregating information from neighboring genes. Graph convolutional networks (GCN) can capture such neighborhood dependency in a graph. However, it is still challenging to apply GCN for SL prediction as SL interactions are extremely sparse, which is more likely to cause overfitting.

RESULTS

In this article, we propose a novel dual-dropout GCN (DDGCN) for learning more robust gene representations for SL prediction. We employ both coarse-grained node dropout and fine-grained edge dropout to address the issue that standard dropout in vanilla GCN is often inadequate in reducing overfitting on sparse graphs. In particular, coarse-grained node dropout can efficiently and systematically enforce dropout at the node (gene) level, while fine-grained edge dropout can further fine-tune the dropout at the interaction (edge) level. We further present a theoretical framework to justify our model architecture. Finally, we conduct extensive experiments on human SL datasets and the results demonstrate the superior performance of our model in comparison with state-of-the-art methods.

AVAILABILITY AND IMPLEMENTATION

DDGCN is implemented in Python 3.7, open-source and freely available at https://github.com/CXX1113/Dual-DropoutGCN.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Matrix (factorization) reloaded: flexible methods for imputing genetic interactions with cross-species and side information

MOTIVATION

Mapping genetic interactions (GIs) can reveal important insights into cellular function and has potential translational applications. There has been great progress in developing high-throughput experimental systems for measuring GIs (e.g. with double knockouts) as well as in defining computational methods for inferring (imputing) unknown interactions. However, existing computational methods for imputation have largely been developed for and applied in baker's yeast, even as experimental systems have begun to allow measurements in other contexts. Importantly, existing methods face a number of limitations in requiring specific side information and with respect to computational cost. Further, few have addressed how GIs can be imputed when data are scarce.

RESULTS

In this article, we address these limitations by presenting a new imputation framework, called Extensible Matrix Factorization (EMF). EMF is a framework of composable models that flexibly exploit cross-species information in the form of GI data across multiple species, and arbitrary side information in the form of kernels (e.g. from protein-protein interaction networks). We perform a rigorous set of experiments on these models in matched GI datasets from baker's and fission yeast. These include the first such experiments on genome-scale GI datasets in multiple species in the same study. We find that EMF models that exploit side and cross-species information improve imputation, especially in data-scarce settings. Further, we show that EMF outperforms the state-of-the-art deep learning method, even when using strictly less data, and incurs orders of magnitude less computational cost.

AVAILABILITY

Implementations of models and experiments are available at: https://github.com/lrgr/EMF.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

SL2MF: Predicting Synthetic Lethality in Human Cancers via Logistic Matrix Factorization

Synthetic lethality (SL) is a promising concept for novel discovery of anti-cancer drug targets. However, wet-lab experiments for detecting SLs are faced with various challenges, such as high cost, low consistency across platforms, or cell lines. Therefore, computational prediction methods are needed to address these issues. This paper proposes a novel SL prediction method, named SL² MF, which employs logistic matrix factorization to learn latent representations of genes from the observed SL data. The probability that two genes are likely to form SL is modeled by the linear combination of gene latent vectors. As known SL pairs are more trustworthy than unknown pairs, we design importance weighting schemes to assign higher importance weights for known SL pairs and lower importance weights for unknown pairs in SL² MF. Moreover, we also incorporate biological knowledge about genes from protein-protein interaction (PPI) data and Gene Ontology (GO). In particular, we calculate the similarity between genes based on their GO annotations and topological properties in the PPI network. Extensive experiments on the SL interaction data from SynLethDB database have been conducted to demonstrate the effectiveness of SL² MF.

EXP2SL: A Machine Learning Framework for Cell-Line-Specific Synthetic Lethality Prediction

Synthetic lethality (SL), an important type of genetic interaction, can provide useful insight into the target identification process for the development of anticancer therapeutics. Although several well-established SL gene pairs have been verified to be conserved in humans, most SL interactions remain cell-line specific. Here, we demonstrated that the cell-line-specific gene expression profiles derived from the shRNA perturbation experiments performed in the LINCS L1000 project can provide useful features for predicting SL interactions in human. In this paper, we developed a semi-supervised neural network-based method called EXP2SL to accurately identify SL interactions from the L1000 gene expression profiles. Through a systematic evaluation on the SL datasets of three different cell lines, we demonstrated that our model achieved better performance than the baseline methods and verified the effectiveness of using the L1000 gene expression features and the semi-supervise training technique in SL prediction.

Functional protein representations from biological networks enable diverse cross-species inference

Transferring knowledge between species is key for many biological applications, but is complicated by divergent and convergent evolution. Many current approaches for this problem leverage sequence and interaction network data to transfer knowledge across species, exemplified by network alignment methods. While these techniques do well, they are limited in scope, creating metrics to address one specific problem or task. We take a different approach by creating an environment where multiple knowledge transfer tasks can be performed using the same protein representations. Specifically, our kernel-based method, MUNK, integrates sequence and network structure to create functional protein representations, embedding proteins from different species in the same vector space. First we show proteins in different species that are close in MUNK-space are functionally similar. Next, we use these representations to share knowledge of synthetic lethal interactions between species. Importantly, we find that the results using MUNK-representations are at least as accurate as existing algorithms for these tasks. Finally, we generalize the notion of a phenolog ('orthologous phenotype') to use functionally similar proteins (i.e. those with similar representations). We demonstrate the utility of this broadened notion by using it to identify known phenologs and novel non-obvious ones supported by current research.

Predicting synthetic lethal interactions using conserved patterns in protein interaction networks

In response to a need for improved treatments, a number of promising novel targeted cancer therapies are being developed that exploit human synthetic lethal interactions. This is facilitating personalised medicine strategies in cancers where specific tumour suppressors have become inactivated. Mainly due to the constraints of the experimental procedures, relatively few human synthetic lethal interactions have been identified. Here we describe SLant (Synthetic Lethal analysis via Network topology), a computational systems approach to predicting human synthetic lethal interactions that works by identifying and exploiting conserved patterns in protein interaction network topology both within and across species. SLant out-performs previous attempts to classify human SSL interactions and experimental validation of the models predictions suggests it may provide useful guidance for future SSL screenings and ultimately aid targeted cancer therapy development.

Data mining for mutation-specific targets in acute myeloid leukemia

Three mutation-specific targeted therapies have recently been approved by the FDA for the treatment of acute myeloid leukemia (AML): midostaurin for FLT3 mutations, enasidenib for relapsed or refractory cases with IDH2 mutations, and ivosidenib for cases with an IDH1 mutation. Together, these agents offer a mutation-directed treatment approach for up to 45% of de novo adult AML cases, a welcome deluge after a prolonged drought. At the same time, a number of computational tools have recently been developed that promise to further accelerate progress in mutation-specific therapy for AML and other cancers. Technical advances together with comprehensively annotated AML tissue banks have resulted in the availability of large and complex data sets for exploration by the end-user, including (i) microarray gene expression, (ii) exome sequencing, (iii) deep sequencing data of sub-clone heterogeneity, (iv) RNA sequencing of gene expression (bulk and single cell), (v) DNA methylation and chromatin, (vi) and germline quantitative trait loci. Yet few clinicians or experimental hematologists have the time or the training to access or analyze these repositories. This review summarizes the data sets and bioinformatic tools currently available to further the discovery of mutation-specific targets with an emphasis on web-based applications that are open, accessible, user-friendly, and do not require coding experience to navigate. We show examples of how available data can be mined to identify potential targets using synthetic lethality, drug repurposing, epigenetic sub-grouping, and proteomic networks while also highlighting strengths and limitations and the need for superior models for validation.

[Searching for essential genes in cancer genomes]

The concept of essential genes, whose loss of functionality leads to cell death, is one of the fundamental concepts of genetics and is important for fundamental and applied research. This field is particularly promising in relation to oncology, since the search for genetic vulnerabilities of cancer cells allows us to identify new potential targets for antitumor therapy. The modern biotechnology capacities allow carrying out large-scale projects for sequencing somatic mutations in tumors, as well as directly interfering the genetic apparatus of cancer cells. They provided accumulation of a considerable body of knowledge about genetic variants and corresponding phenotypic manifestations in tumors. In the near future this knowledge will find application in clinical practice. This review describes the main experimental and computational approaches to the search for essential genes, concentrating on the application of these methods in the field of molecular oncology.

Computational Approaches to Identify Genetic Interactions for Cancer Therapeutics

The development of improved cancer therapies is frequently cited as an urgent unmet medical need. Here we describe how genetic interactions are being therapeutically exploited to identify novel targeted treatments for cancer. We discuss the current methodologies that use 'omics data to identify genetic interactions, in particular focusing on synthetic sickness lethality (SSL) and synthetic dosage lethality (SDL). We describe the experimental and computational approaches undertaken both in humans and model organisms to identify these interactions. Finally we discuss some of the identified targets with licensed drugs, inhibitors in clinical trials or with compounds under development.

Systematic discovery of mutation-specific synthetic lethals by mining pan-cancer human primary tumor data

Two genes are synthetically lethal (SL) when defects in both are lethal to a cell but a single defect is non-lethal. SL partners of cancer mutations are of great interest as pharmacological targets; however, identifying them by cell line-based methods is challenging. Here we develop MiSL (Mining Synthetic Lethals), an algorithm that mines pan-cancer human primary tumour data to identify mutation-specific SL partners for specific cancers. We apply MiSL to 12 different cancers and predict 145,891 SL partners for 3,120 mutations, including known mutation-specific SL partners. Comparisons with functional screens show that MiSL predictions are enriched for SLs in multiple cancers. We extensively validate a SL interaction identified by MiSL between the IDH1 mutation and ACACA in leukaemia using gene targeting and patient-derived xenografts. Furthermore, we apply MiSL to pinpoint genetic biomarkers for drug sensitivity. These results demonstrate that MiSL can accelerate precision oncology by identifying mutation-specific targets and biomarkers.

There are no robust methods for systematically identifying mutation-specific synthetic lethal (SL) partners in cancer. Here, the authors develop a computational algorithm that uses pan-cancer data to detect mutation-andcancer-specific SL partners and they validate a novel SL interaction between mutant IDH and loss of ACACA in leukaemia.

'Big data' approaches for novel anti-cancer drug discovery

INTRODUCTION

The development of improved cancer therapies is frequently cited as an urgent unmet medical need. Recent advances in platform technologies and the increasing availability of biological 'big data' are providing an unparalleled opportunity to systematically identify the key genes and pathways involved in tumorigenesis. The discoveries made using these new technologies may lead to novel therapeutic interventions. Areas covered: The authors discuss the current approaches that use 'big data' to identify cancer drivers. These approaches include the analysis of genomic sequencing data, pathway data, multi-platform data, identifying genetic interactions such as synthetic lethality and using cell line data. They review how big data is being used to identify novel drug targets. The authors then provide an overview of the available data repositories and tools being used at the forefront of cancer drug discovery. Expert opinion: Targeted therapies based on the genomic events driving the tumour will eventually inform treatment protocols. However, using a tailored approach to treat all tumour patients may require developing a large repertoire of targeted drugs.