Comprehensive anticancer drug response prediction based on a simple cell line-drug complex network model

Singular value thresholding two-stage matrix completion for drug sensitivity discovery

Incomplete data presents significant challenges in drug sensitivity analysis, especially in critical areas like oncology, where precision is paramount. Our study introduces an innovative imputation method designed specifically for low-rank matrices, addressing the crucial challenge of data completion in anticancer drug sensitivity testing. Our method unfolds in two main stages: Initially, the singular value thresholding algorithm is employed for preliminary matrix completion, establishing a solid foundation for subsequent steps. Then, the matrix rows are segmented into distinct blocks based on hierarchical clustering of correlation coefficients, applying singular value thresholding to the largest block, which has been proved to possess the largest entropy. This is followed by a refined data restoration process, where the reconstructed largest block is integrated into the initial matrix completion to achieve the final matrix completion. Compared to other methods, our approach not only improves the accuracy of data restoration but also ensures the integrity and reliability of the imputed values, establishing it as a robust tool for future drug sensitivity analysis.

Learning the cellular activity representation based on gene regulatory networks for prediction of tumor response to drugs

Predicting the response of tumor cells to anti-tumor drugs is critical to realizing cancer precision medicine. Currently, most existing methods ignore the regulatory relationships between genes and thus have unsatisfactory predictive performance. In this paper, we propose to predict anti-tumor drug efficacy via learning the activity representation of tumor cells based on a priori knowledge of gene regulation networks (GRNs). Specifically, the method simulates the cellular biosystem by synthesizing a cell-gene activity network and then infers a new low-dimensional activity representation for tumor cells from the raw high-dimensional expression profile. The simulated cell-gene network mainly comprises known gene regulatory networks collected from multiple resources and fuses tumor cells by linking them to hotspot genes that are over- or under-expressed in them. The resulting activity representation could not only reflect the shallow expression profile (hotspot genes) but also mines in-depth information of gene regulation activity in tumor cells before treatment. Finally, we build deep learning models on the activity representation for predicting drug efficacy in tumor cells. Experimental results on the benchmark GDSC dataset demonstrate the superior performance of the proposed method over SOTA methods with the highest AUC of 0.954 in the efficacy label prediction and the best R2 of 0.834 in the regression of half maximal inhibitory concentration (IC50) values, suggesting the potential value of the proposed method in practice.

SNSynergy: Similarity network-based machine learning framework for synergy prediction towards new cell lines and new anticancer drug combinations

The computational method has been proven to be a promising means for pre-screening large-scale anticancer drug combinations to support precision oncology applications. Pioneering efforts have been made to develop machine learning technology for predicting drug synergy, but high computational cost for training models as well as great diversity and limited size in screening data escalate the difficulty of prediction. To address this challenge, we propose a simple machine learning framework, namely Similarity Network-based Synergy prediction (SNSynergy), for predicting synergistic effects towards new cell lines and new drug combinations by two locally weighted models CLSN and DCSN. This framework only requires a small amount of auxiliary data, like genomics information of cell lines and the molecular fingerprints or targets of drugs. Based on the assumption that similar cell lines and similar drug combinations have similar synergistic effects, CLSN and DCSN predict synergy scores through capturing individual synergy contributions of nearest cell line and drug combination neighbors, respectively. High correlations between predicted and measured synergy scores on two leading cancer cell line pharmacogenomic screening datasets (the O'Neil dataset and the NCI-ALMANAC dataset) demonstrate the effectiveness and robustness of SNSynergy. Many of the identified drug combinations are consistent with previous studies, or have been explored in clinical settings against the specific cancer type, showing that SNSynergy has the potential to supply cost-saving and effective high-throughput screening for prioritizing the most applicable cell lines and the most promising drug combinations.

StringFix: an annotation-guided transcriptome assembler improves the recovery of amino acid sequences from RNA-Seq reads

BACKGROUND

Reconstruction of amino acid sequences from assembled transcriptome is of interest in personalized medicine, for example, to predict drug-target (or protein-protein) interaction considering individual's genomic variations. Most of the existing transcriptome assemblers, however, seems not well suited for this purpose.

METHODS

In this work, we present StringFix, an annotation guided transcriptome assembly and protein sequence reconstruction software tool that takes genome-aligned reads and the annotations associated to the reference genome as input. The tool 'fixes' the pre-annotated transcript sequence by taking small variations into account, finally to produce possible amino acid sequences that are likely to exist in the test tissue.

RESULTS

The results show that, using outputs from existing reference-based assemblers as the input GTF-guide, StringFix could reconstruct amino acid sequences more precisely with higher sensitivity than direct generation using the recovered transcripts from all the assemblers we tested.

CONCLUSION

By using StringFix with the existing reference-based assemblers, one can recover not only a novel transcripts and isoforms but also the possible amino acid sequence stemming from them.

A performance evaluation of drug response prediction models for individual drugs

Drug response prediction is important to establish personalized medicine for cancer therapy. Model construction for predicting drug response (i.e., cell viability half-maximal inhibitory concentration [IC₅₀]) of an individual drug by inputting pharmacogenomics in disease models remains critical. Machine learning (ML) has been predominantly applied for prediction, despite the advent of deep learning (DL). Moreover, whether DL or traditional ML models are superior for predicting cell viability IC_50s has to be established. Herein, we constructed ML and DL drug response prediction models for 24 individual drugs and compared the performance of the models by employing gene expression and mutation profiles of cancer cell lines as input. We observed no significant difference in drug response prediction performance between DL and ML models for 24 drugs [root mean squared error (RMSE) ranging from 0.284 to 3.563 for DL and from 0.274 to 2.697 for ML; R² ranging from -7.405 to 0.331 for DL and from -8.113 to 0.470 for ML]. Among the 24 individual drugs, the ridge model of panobinostat exhibited the best performance (R² 0.470 and RMSE 0.623). Thus, we selected the ridge model of panobinostat for further application of explainable artificial intelligence (XAI). Using XAI, we further identified important genomic features for panobinostat response prediction in the ridge model, suggesting the genomic features of 22 genes. Based on our findings, results for an individual drug employing both DL and ML models were comparable. Our study confirms the applicability of drug response prediction models for individual drugs.

Drug mechanism enrichment analysis improves prioritization of therapeutics for repurposing

BACKGROUND

There is a pressing need for improved methods to identify effective therapeutics for diseases. Many computational approaches have been developed to repurpose existing drugs to meet this need. However, these tools often output long lists of candidate drugs that are difficult to interpret, and individual drug candidates may suffer from unknown off-target effects. We reasoned that an approach which aggregates information from multiple drugs that share a common mechanism of action (MOA) would increase on-target signal compared to evaluating drugs on an individual basis. In this study, we present drug mechanism enrichment analysis (DMEA), an adaptation of gene set enrichment analysis (GSEA), which groups drugs with shared MOAs to improve the prioritization of drug repurposing candidates.

RESULTS

First, we tested DMEA on simulated data and showed that it can sensitively and robustly identify an enriched drug MOA. Next, we used DMEA on three types of rank-ordered drug lists: (1) perturbagen signatures based on gene expression data, (2) drug sensitivity scores based on high-throughput cancer cell line screening, and (3) molecular classification scores of intrinsic and acquired drug resistance. In each case, DMEA detected the expected MOA as well as other relevant MOAs. Furthermore, the rankings of MOAs generated by DMEA were better than the original single-drug rankings in all tested data sets. Finally, in a drug discovery experiment, we identified potential senescence-inducing and senolytic drug MOAs for primary human mammary epithelial cells and then experimentally validated the senolytic effects of EGFR inhibitors.

CONCLUSIONS

DMEA is a versatile bioinformatic tool that can improve the prioritization of candidates for drug repurposing. By grouping drugs with a shared MOA, DMEA increases on-target signal and reduces off-target effects compared to analysis of individual drugs. DMEA is publicly available as both a web application and an R package at https://belindabgarana.github.io/DMEA .

CTDN (Convolutional Temporal Based Deep- Neural Network): An Improvised Stacked Hybrid Computational Approach for Anticancer Drug Response Prediction

The characterization of drug - metabolizing enzymes is a significant problem for customized therapy. It is important to choose the right drugs for cancer victims, and the ability to forecast how those drugs will react is usually based on the available information, genetic sequence, and structural properties. To the finest of our knowledge, this is the first study to evaluate optimization algorithms for selection of features and pharmacogenetics categorization using classification methods based on a successful evolutionary algorithm using datasets from the Cancer Cell Line Encyclopaedia (CCLE) and Genomics of Drug Sensitivity in Cancer (GDSC). The study proposes the uses of Firefly and Grey Wolf Optimization techniques for feature extraction, while comparing the traditional Machine Learning (ML), ensemble ML and Stacking Algorithm with the proposed Convolutional Temporal Deep Neural Network or CTDN. With the potential to increase efficiency from the suggested intelligible classifier model for a suggestive chemotherapeutic drugs response prediction, our study is important in particular for selecting an acceptable feature selection method. The comparison analysis demonstrates that the proposed model not only surpasses the prior state-of-the-art methods, but also uses Grey Wolf and Fire Fly Optimization to lessen multicollinearity and overfitting.

Gamma distribution based predicting model for breast cancer drug response based on multi-layer feature selection

In the pursuit of precision medicine for cancer, a promising step is to predict drug response based on data mining, which can provide clinical decision support for cancer patients. Although some machine learning methods for predicting drug response from genomic data already exist, most of them focus on point prediction, which cannot reveal the distribution of predicted results. In this paper, we propose a three-layer feature selection combined with a gamma distribution based GLM and a two-layer feature selection combined with an ANN. The two regression methods are applied to the Encyclopedia of Cancer Cell Lines (CCLE) and the Cancer Drug Sensitivity Genomics (GDSC) datasets. Using ten-fold cross-validation, our methods achieve higher accuracy on anticancer drug response prediction compared to existing methods, with an R ² and RMSE of 0.87 and 0.53, respectively. Through data validation, the significance of assessing the reliability of predictions by predicting confidence intervals and its role in personalized medicine are illustrated. The correlation analysis of the genes selected from the three layers of features also shows the effectiveness of our proposed methods.

Low-rank regression models for multiple binary responses and their applications to cancer cell-line encyclopedia data

In this paper, we study high-dimensional multivariate logistic regression models in which a common set of covariates is used to predict multiple binary outcomes simultaneously. Our work is primarily motivated from many biomedical studies with correlated multiple responses such as the cancer cell-line encyclopedia project. We assume that the underlying regression coefficient matrix is simultaneously low-rank and row-wise sparse. We propose an intuitively appealing selection and estimation framework based on marginal model likelihood, and we develop an efficient computational algorithm for inference. We establish a novel high-dimensional theory for this nonlinear multivariate regression. Our theory is general, allowing for potential correlations between the binary responses. We propose a new type of nuclear norm penalty using the smooth clipped absolute deviation, filling the gap in the related non-convex penalization literature. We theoretically demonstrate that the proposed approach improves estimation accuracy by considering multiple responses jointly through the proposed estimator when the underlying coefficient matrix is low-rank and row-wise sparse. In particular, we establish the non-asymptotic error bounds, and both rank and row support consistency of the proposed method. Moreover, we develop a consistent rule to simultaneously select the rank and row dimension of the coefficient matrix. Furthermore, we extend the proposed methods and theory to a joint Ising model, which accounts for the dependence relationships. In our analysis of both simulated data and the cancer cell line encyclopedia data, the proposed methods outperform the existing methods in better predicting responses.

PredictiveNetwork: predictive gene network estimation with application to gastric cancer drug response-predictive network analysis

Background

Gene regulatory networks have garnered a large amount of attention to understand disease mechanisms caused by complex molecular network interactions. These networks have been applied to predict specific clinical characteristics, e.g., cancer, pathogenicity, and anti-cancer drug sensitivity. However, in most previous studies using network-based prediction, the gene networks were estimated first, and predicted clinical characteristics based on pre-estimated networks. Thus, the estimated networks cannot describe clinical characteristic-specific gene regulatory systems. Furthermore, existing computational methods were developed from algorithmic and mathematics viewpoints, without considering network biology.

Results

To effectively predict clinical characteristics and estimate gene networks that provide critical insights into understanding the biological mechanisms involved in a clinical characteristic, we propose a novel strategy for predictive gene network estimation. The proposed strategy simultaneously performs gene network estimation and prediction of the clinical characteristic. In this strategy, the gene network is estimated with minimal network estimation and prediction errors. We incorporate network biology by assuming that neighboring genes in a network have similar biological functions, while hub genes play key roles in biological processes. Thus, the proposed method provides interpretable prediction results and enables us to uncover biologically reliable marker identification. Monte Carlo simulations shows the effectiveness of our method for feature selection in gene estimation and prediction with excellent prediction accuracy. We applied the proposed strategy to construct gastric cancer drug-responsive networks.

Conclusion

We identified gastric drug response predictive markers and drug sensitivity/resistance-specific markers, AKR1B10, AKR1C3, ANXA10, and ZNF165, based on GDSC data analysis. Our results for identifying drug sensitive and resistant specific molecular interplay are strongly supported by previous studies. We expect that the proposed strategy will be a useful tool for uncovering crucial molecular interactions involved a specific biological mechanism, such as cancer progression or acquired drug resistance.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-022-04871-z.

Network approaches for modeling the effect of drugs and diseases

The network approach is quickly becoming a fundamental building block of computational methods aiming at elucidating the mechanism of action (MoA) and therapeutic effect of drugs. By modeling the effect of drugs and diseases on different biological networks, it is possible to better explain the interplay between disease perturbations and drug targets as well as how drug compounds induce favorable biological responses and/or adverse effects. Omics technologies have been extensively used to generate the data needed to study the mechanisms of action of drugs and diseases. These data are often exploited to define condition-specific networks and to study whether drugs can reverse disease perturbations. In this review, we describe network data mining algorithms that are commonly used to study drug’s MoA and to improve our understanding of the basis of chronic diseases. These methods can support fundamental stages of the drug development process, including the identification of putative drug targets, the in silico screening of drug compounds and drug combinations for the treatment of diseases. We also discuss recent studies using biological and omics-driven networks to search for possible repurposed FDA-approved drug treatments for SARS-CoV-2 infections (COVID-19).

Xprediction: Explainable EGFR-TKIs response prediction based on drug sensitivity specific gene networks

In recent years, drug sensitivity prediction has garnered a great deal of attention due to the growing interest in precision medicine. Several computational methods have been developed for drug sensitivity prediction and the identification of related markers. However, most previous studies have ignored genetic interaction, although complex diseases (e.g., cancer) involve many genes intricately connected in a molecular network rather than the abnormality of a single gene. To effectively predict drug sensitivity and understand its mechanism, we propose a novel strategy for explainable drug sensitivity prediction based on sample-specific gene regulatory networks, designated Xprediction. Our strategy first estimates sample-specific gene regulatory networks that enable us to identify the molecular interplay underlying varying clinical characteristics of cell lines. We then, predict drug sensitivity based on the estimated sample-specific gene regulatory networks. The predictive models are based on machine learning approaches, i.e., random forest, kernel support vector machine, and deep neural network. Although the machine learning models provide remarkable results for prediction and classification, we cannot understand how the models reach their decisions. In other words, the methods suffer from the black box problem and thus, we cannot identify crucial molecular interactions that involve drug sensitivity-related mechanisms. To address this issue, we propose a method that describes the importance of each molecular interaction for the drug sensitivity prediction result. The proposed method enables us to identify crucial gene-gene interactions and thereby, interpret the prediction results based on the identified markers. To evaluate our strategy, we applied Xprediction to EGFR-TKIs prediction based on drug sensitivity specific gene regulatory networks and identified important molecular interactions for EGFR-TKIs prediction. Our strategy effectively performed drug sensitivity prediction compared with prediction based on the expression levels of genes. We also verified through literature, the EGFR-TKIs-related mechanisms of a majority of the identified markers. We expect our strategy to be a useful tool for predicting tasks and uncovering complex mechanisms related to pharmacological profiles, such as mechanisms of acquired drug resistance or sensitivity of cancer cells.

A comprehensive evaluation of regression-based drug responsiveness prediction models, using cell viability inhibitory concentrations (IC50 values)

MOTIVATION

Predicting drug response is critical for precision medicine. Diverse methods have predicted drug responsiveness, as measured by the half-maximal drug inhibitory concentration (IC50), in cultured cells. Although IC50s are continuous, traditional prediction models have dealt mainly with binary classification of responsiveness. However, since there are few regression-based IC50 predictions, comprehensive evaluations of regression-based IC50 prediction models, including machine learning (ML) and deep learning (DL), for diverse data types and dataset sizes, have not been addressed.

RESULTS

Here, we constructed 11 input data settings, including multi-omics settings, with varying dataset sizes, then evaluated the performance of regression-based ML and DL models to predict IC50s. DL models considered two convolutional neural network architectures: CDRScan and residual neural network (ResNet). ResNet was introduced in regression-based DL models for predicting drug response for the first time. As a result, DL models performed better than ML models in all the settings. Also, ResNet performed better than or comparable to CDRScan and ML models in all settings.

AVAILABILITY AND IMPLEMENTATION

The data underlying this article are available in GitHub at https://github.com/labnams/IC50evaluation.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Pan-Cancer Prediction of Cell-Line Drug Sensitivity Using Network-Based Methods

The development of reliable predictive models for individual cancer cell lines to identify an optimal cancer drug is a crucial step to accelerate personalized medicine, but vast differences in cancer cell lines and drug characteristics make it quite challenging to develop predictive models that result in high predictive power and explain the similarity of cell lines or drugs. Our study proposes a novel network-based methodology that breaks the problem into smaller, more interpretable problems to improve the predictive power of anti-cancer drug responses in cell lines. For the drug-sensitivity study, we used the GDSC database for 915 cell lines and 200 drugs. The theory of optimal mass transport was first used to separately cluster cell lines and drugs, using gene-expression profiles and extensive cheminformatic drug features, represented in a form of data networks. To predict cell-line specific drug responses, random forest regression modeling was separately performed for each cell-line drug cluster pair. Post-modeling biological analysis was further performed to identify potential biological correlates associated with drug responses. The network-based clustering method resulted in 30 distinct cell-line drug cluster pairs. Predictive modeling on each cell-line-drug cluster outperformed alternative computational methods in predicting drug responses. We found that among the four drugs top-ranked with respect to prediction performance, three targeted the PI3K/mTOR signaling pathway. Predictive modeling on clustered subsets of cell lines and drugs improved the prediction accuracy of cell-line specific drug responses. Post-modeling analysis identified plausible biological processes associated with drug responses.

An overview of machine learning methods for monotherapy drug response prediction

For an increasing number of preclinical samples, both detailed molecular profiles and their responses to various drugs are becoming available. Efforts to understand, and predict, drug responses in a data-driven manner have led to a proliferation of machine learning (ML) methods, with the longer term ambition of predicting clinical drug responses. Here, we provide a uniquely wide and deep systematic review of the rapidly evolving literature on monotherapy drug response prediction, with a systematic characterization and classification that comprises more than 70 ML methods in 13 subclasses, their input and output data types, modes of evaluation, and code and software availability. ML experts are provided with a fundamental understanding of the biological problem, and how ML methods are configured for it. Biologists and biomedical researchers are introduced to the basic principles of applicable ML methods, and their application to the problem of drug response prediction. We also provide systematic overviews of commonly used data sources used for training and evaluation methods.

Dissecting the Genome for Drug Response Prediction

The prediction of the cancer cell lines sensitivity to a specific treatment is one of the current challenges in precision medicine. With omics and pharmacogenomics data being available for over 1000 cancer cell lines, several machine learning and deep learning algorithms have been proposed for drug sensitivity prediction. However, deciding which omics data to use and which computational methods can efficiently incorporate data from different sources is the challenge which several research groups are working on. In this review, we summarize recent advances in the representative computational methods that have been developed in the last 2 years on three public datasets: COSMIC, CCLE, NCI-60. These methods aim to improve the prediction of the cancer cell lines sensitivity to a given treatment by incorporating drug's chemical information in the input or using a priori feature selection. Finally, we discuss the latest published method which aims to improve the prediction of clinical drug response of real patients starting from cancer cell line molecular profiles.

DNA Repair Genes as Drug Candidates for Early Breast Cancer Onset in Latin America: A Systematic Review

The prevalence of breast cancer in young women (YWBC) has increased alarmingly. Significant efforts are being made to elucidate the biological mechanisms concerning the development, prognosis, and pathological response in early-onset breast cancer (BC) patients. Dysfunctional DNA repair proteins are implied in BC predisposition, progression, and therapy response, underscoring the need for further analyses on DNA repair genes. Public databases of large patient datasets such as METABRIC, TCGA, COSMIC, and cancer cell lines allow the identification of variants in DNA repair genes and possible precision drug candidates. This study aimed at identifying variants and drug candidates that may benefit Latin American (LA) YWBC. We analyzed pathogenic variants in 90 genes involved in DNA repair in public BC datasets from METABRIC, TCGA, COSMIC, CCLE, and COSMIC Cell Lines Project. Results showed that reported DNA repair germline variants in the LA dataset are underrepresented in large databases, in contrast to other populations. Additionally, only six gene repair variants in women under 50 years old from the study population were reported in BC cell lines. Therefore, there is a need for new approaches to study DNA repair variants reported in young women from LA.

Predicting Anti-Cancer Drug Response by Finding Optimal Subset of Drugs

MOTIVATION

One of the most difficult challenges in precision medicine is determining the best treatment strategy for each patient based on personal information. Since drug response prediction in vitro is extremely expensive, time-consuming, and virtually impossible, and because there are so many cell lines and drug data, computational methods are needed.

RESULTS

MinDrug is a method for predicting anti-cancer drug response which try to identify the best subset of drugs that are the most similar to other drugs. MinDrug predicts the anti-cancer drug response on a new cell line using information from drugs in this subset and their connections to other drugs. MinDrug employs a heuristic star algorithm to identify an optimal subset of drugs and a regression technique known as Elastic-Net approaches to predict anti-cancer drug response in a new cell line. To test MinDrug, we use both statistical and biological methods to assess the selected drugs. MinDrug is also compared to four state-of-the-art approaches using various k-fold cross-validations on two large public datasets: GDSC and CCLE. MinDrug outperforms the other approaches in terms of precision, robustness, and speed. Furthermore, we compare the evaluation results of all the approaches with an external dataset with a statistical distribution that is not exactly the same as the training data. The results show that MinDrug continues to outperform the other approaches.

AVAILABILITY

MinDrug's source code can be found at https://github.com/yassaee/MinDrug.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Predicting breast cancer drug response using a multiple-layer cell line drug response network model

Background

Predicting patient drug response based on a patient’s molecular profile is one of the key goals of precision medicine in breast cancer (BC). Multiple drug response prediction models have been developed to address this problem. However, most of them were developed to make sensitivity predictions for multiple single drugs within cell lines from various cancer types instead of a single cancer type, do not take into account drug properties, and have not been validated in cancer patient-derived data. Among the multi-omics data, gene expression profiles have been shown to be the most informative data for drug response prediction. However, these models were often developed with individual genes. Therefore, this study aimed to develop a drug response prediction model for BC using multiple data types from both cell lines and drugs.

Methods

We first collected the baseline gene expression profiles of 49 BC cell lines along with IC₅₀ values for 220 drugs tested in these cell lines from Genomics of Drug Sensitivity in Cancer (GDSC). Using these data, we developed a multiple-layer cell line-drug response network (ML-CDN2) by integrating a one-layer cell line similarity network based on the pathway activity profiles and a three-layer drug similarity network based on the drug structures, targets, and pan-cancer IC₅₀ profiles. We further used ML-CDN2 to predict the drug response for new BC cell lines or patient-derived samples.

Results

ML-CDN2 demonstrated a good predictive performance, with the Pearson correlation coefficient between the observed and predicted IC₅₀ values for all GDSC cell line-drug pairs of 0.873. Also, ML-CDN2 showed a good performance when used to predict drug response in new BC cell lines from the Cancer Cell Line Encyclopedia (CCLE), with a Pearson correlation coefficient of 0.718. Moreover, we found that the cell line-derived ML-CDN2 model could be applied to predict drug response in the BC patient-derived samples from The Cancer Genome Atlas (TCGA).

Conclusions

The ML-CDN2 model was built to predict BC drug response using comprehensive information from both cell lines and drugs. Compared with existing methods, it has the potential to predict the drug response for BC patient-derived samples.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-021-08359-6.

Current Advances and Limitations of Deep Learning in Anticancer Drug Sensitivity Prediction

Anticancer drug screening can accelerate drug discovery to save the lives of cancer patients, but cancer heterogeneity makes this screening challenging. The prediction of anticancer drug sensitivity is useful for anticancer drug development and the identification of biomarkers of drug sensitivity. Deep learning, as a branch of machine learning, is an important aspect of in silico research. Its outstanding computational performance means that it has been used for many biomedical purposes, such as medical image interpretation, biological sequence analysis, and drug discovery. Several studies have predicted anticancer drug sensitivity based on deep learning algorithms. The field of deep learning has made progress regarding model performance and multi-omics data integration. However, deep learning is limited by the number of studies performed and data sources available, so it is not perfect as a pre-clinical approach for use in the anticancer drug screening process. Improving the performance of deep learning models is a pressing issue for researchers. In this review, we introduce the research of anticancer drug sensitivity prediction and the use of deep learning in this research area. To provide a reference for future research, we also review some common data sources and machine learning methods. Lastly, we discuss the advantages and disadvantages of deep learning, as well as the limitations and future perspectives regarding this approach.

A computational method for drug sensitivity prediction of cancer cell lines based on various molecular information

Determining sensitive drugs for a patient is one of the most critical problems in precision medicine. Using genomic profiles of the tumor and drug information can help in tailoring the most efficient treatment for a patient. In this paper, we proposed a classification machine learning approach that predicts the sensitive/resistant drugs for a cell line. It can be performed by using both drug and cell line similarities, one of the cell line or drug similarities, or even not using any similarity information. This paper investigates the influence of using previously defined as well as two newly introduced similarities on predicting anti-cancer drug sensitivity. The proposed method uses max concentration thresholds for assigning drug responses to class labels. Its performance was evaluated using stratified five-fold cross-validation on cell line-drug pairs in two datasets. Assessing the predictive powers of the proposed model and three sets of methods, including state-of-the-art classification methods, state-of-the-art regression methods, and off-the-shelf classification machine learning approaches shows that the proposed method outperforms other methods. Moreover, The efficiency of the model is evaluated in tissue-specific conditions. Besides, the novel sensitive associations predicted by this model were verified by several supportive evidence in the literature and reliable database. Therefore, the proposed model can efficiently be used in predicting anti-cancer drug sensitivity. Material and implementation are available at https://github.com/fahmadimoughari/CDSML.

Predicting tumor response to drugs based on gene-expression biomarkers of sensitivity learned from cancer cell lines

Background

Human cancer cell line profiling and drug sensitivity studies provide valuable information about the therapeutic potential of drugs and their possible mechanisms of action. The goal of those studies is to translate the findings from in vitro studies of cancer cell lines into in vivo therapeutic relevance and, eventually, patients’ care. Tremendous progress has been made.

Results

In this work, we built predictive models for 453 drugs using data on gene expression and drug sensitivity (IC₅₀) from cancer cell lines. We identified many known drug-gene interactions and uncovered several potentially novel drug-gene associations. Importantly, we further applied these predictive models to ~ 17,000 bulk RNA-seq samples from The Cancer Genome Atlas (TCGA) and the Genotype-Tissue Expression (GTEx) database to predict drug sensitivity for both normal and tumor tissues. We created a web site for users to visualize and download our predicted data (https://manticore.niehs.nih.gov/cancerRxTissue). Using trametinib as an example, we showed that our approach can faithfully recapitulate the known tumor specificity of the drug.

Conclusions

We demonstrated that our approach can predict drugs that 1) are tumor-type specific; 2) elicit higher sensitivity from tumor compared to corresponding normal tissue; 3) elicit differential sensitivity across breast cancer subtypes. If validated, our prediction could have relevance for preclinical drug testing and in phase I clinical design.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-021-07581-7.

Machine learning based anti-cancer drug response prediction and search for predictor genes using cancer cell line gene expression

Although many models have been proposed to accurately predict the response of drugs in cell lines recent years, understanding the genome related to drug response is also the key for completing oncology precision medicine. In this paper, based on the cancer cell line gene expression and the drug response data, we established a reliable and accurate drug response prediction model and found predictor genes for some drugs of interest. To this end, we first performed pre-selection of genes based on the Pearson correlation coefficient and then used ElasticNet regression model for drug response prediction and fine gene selection. To find more reliable set of predictor genes, we performed regression twice for each drug, one with IC₅₀ and the other with area under the curve (AUC) (or activity area). For the 12 drugs we tested, the predictive performance in terms of Pearson correlation coefficient exceeded 0.6 and the highest one was 17-AAG for which Pearson correlation coefficient was 0.811 for IC₅₀ and 0.81 for AUC. We identify common predictor genes for IC₅₀ and AUC, with which the performance was similar to those with genes separately found for IC₅₀ and AUC, but with much smaller number of predictor genes. By using only common predictor genes, the highest performance was AZD6244 (0.8016 for IC₅₀, 0.7945 for AUC) with 321 predictor genes.

Exploration of databases and methods supporting drug repurposing: a comprehensive survey

Drug development involves a deep understanding of the mechanisms of action and possible side effects of each drug, and sometimes results in the identification of new and unexpected uses for drugs, termed as drug repurposing. Both in case of serendipitous observations and systematic mechanistic explorations, confirmation of new indications for a drug requires hypothesis building around relevant drug-related data, such as molecular targets involved, and patient and cellular responses. These datasets are available in public repositories, but apart from sifting through the sheer amount of data imposing computational bottleneck, a major challenge is the difficulty in selecting which databases to use from an increasingly large number of available databases. The database selection is made harder by the lack of an overview of the types of data offered in each database. In order to alleviate these problems and to guide the end user through the drug repurposing efforts, we provide here a survey of 102 of the most promising and drug-relevant databases reported to date. We summarize the target coverage and types of data available in each database and provide several examples of how multi-database exploration can facilitate drug repurposing.

DeepCDR: a hybrid graph convolutional network for predicting cancer drug response

MOTIVATION

Accurate prediction of cancer drug response (CDR) is challenging due to the uncertainty of drug efficacy and heterogeneity of cancer patients. Strong evidences have implicated the high dependence of CDR on tumor genomic and transcriptomic profiles of individual patients. Precise identification of CDR is crucial in both guiding anti-cancer drug design and understanding cancer biology.

RESULTS

In this study, we present DeepCDR which integrates multi-omics profiles of cancer cells and explores intrinsic chemical structures of drugs for predicting CDR. Specifically, DeepCDR is a hybrid graph convolutional network consisting of a uniform graph convolutional network and multiple subnetworks. Unlike prior studies modeling hand-crafted features of drugs, DeepCDR automatically learns the latent representation of topological structures among atoms and bonds of drugs. Extensive experiments showed that DeepCDR outperformed state-of-the-art methods in both classification and regression settings under various data settings. We also evaluated the contribution of different types of omics profiles for assessing drug response. Furthermore, we provided an exploratory strategy for identifying potential cancer-associated genes concerning specific cancer types. Our results highlighted the predictive power of DeepCDR and its potential translational value in guiding disease-specific drug design.

AVAILABILITY AND IMPLEMENTATION

DeepCDR is freely available at https://github.com/kimmo1019/DeepCDR.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Deep learning of pharmacogenomics resources: moving towards precision oncology

The recent accumulation of cancer genomic data provides an opportunity to understand how a tumor's genomic characteristics can affect its responses to drugs. This field, called pharmacogenomics, is a key area in the development of precision oncology. Deep learning (DL) methodology has emerged as a powerful technique to characterize and learn from rapidly accumulating pharmacogenomics data. We introduce the fundamentals and typical model architectures of DL. We review the use of DL in classification of cancers and cancer subtypes (diagnosis and treatment stratification of patients), prediction of drug response and drug synergy for individual tumors (treatment prioritization for a patient), drug repositioning and discovery and the study of mechanism/mode of action of treatments. For each topic, we summarize current genomics and pharmacogenomics data resources such as pan-cancer genomics data for cancer cell lines (CCLs) and tumors, and systematic pharmacologic screens of CCLs. By revisiting the published literature, including our in-house analyses, we demonstrate the unprecedented capability of DL enabled by rapid accumulation of data resources to decipher complex drug response patterns, thus potentially improving cancer medicine. Overall, this review provides an in-depth summary of state-of-the-art DL methods and up-to-date pharmacogenomics resources and future opportunities and challenges to realize the goal of precision oncology.

Quantitative Structure-Mutation-Activity Relationship Tests (QSMART) model for protein kinase inhibitor response prediction

BACKGROUND

Protein kinases are a large family of druggable proteins that are genomically and proteomically altered in many human cancers. Kinase-targeted drugs are emerging as promising avenues for personalized medicine because of the differential response shown by altered kinases to drug treatment in patients and cell-based assays. However, an incomplete understanding of the relationships connecting genome, proteome and drug sensitivity profiles present a major bottleneck in targeting kinases for personalized medicine.

RESULTS

In this study, we propose a multi-component Quantitative Structure-Mutation-Activity Relationship Tests (QSMART) model and neural networks framework for providing explainable models of protein kinase inhibition and drug response ([Formula: see text]) profiles in cell lines. Using non-small cell lung cancer as a case study, we show that interaction terms that capture associations between drugs, pathways, and mutant kinases quantitatively contribute to the response of two EGFR inhibitors (afatinib and lapatinib). In particular, protein-protein interactions associated with the JNK apoptotic pathway, associations between lung development and axon extension, and interaction terms connecting drug substructures and the volume/charge of mutant residues at specific structural locations contribute significantly to the observed [Formula: see text] values in cell-based assays.

CONCLUSIONS

By integrating multi-omics data in the QSMART model, we not only predict drug responses in cancer cell lines with high accuracy but also identify features and explainable interaction terms contributing to the accuracy. Although we have tested our multi-component explainable framework on protein kinase inhibitors, it can be extended across the proteome to investigate the complex relationships connecting genotypes and drug sensitivity profiles.

Leveraging TCGA gene expression data to build predictive models for cancer drug response

BACKGROUND

Machine learning has been utilized to predict cancer drug response from multi-omics data generated from sensitivities of cancer cell lines to different therapeutic compounds. Here, we build machine learning models using gene expression data from patients' primary tumor tissues to predict whether a patient will respond positively or negatively to two chemotherapeutics: 5-Fluorouracil and Gemcitabine.

RESULTS

We focused on 5-Fluorouracil and Gemcitabine because based on our exclusion criteria, they provide the largest numbers of patients within TCGA. Normalized gene expression data were clustered and used as the input features for the study. We used matching clinical trial data to ascertain the response of these patients via multiple classification methods. Multiple clustering and classification methods were compared for prediction accuracy of drug response. Clara and random forest were found to be the best clustering and classification methods, respectively. The results show our models predict with up to 86% accuracy; despite the study's limitation of sample size. We also found the genes most informative for predicting drug response were enriched in well-known cancer signaling pathways and highlighted their potential significance in chemotherapy prognosis.

CONCLUSIONS

Primary tumor gene expression is a good predictor of cancer drug response. Investment in larger datasets containing both patient gene expression and drug response is needed to support future work of machine learning models. Ultimately, such predictive models may aid oncologists with making critical treatment decisions.

ADRML: anticancer drug response prediction using manifold learning

One of the prominent challenges in precision medicine is to select the most appropriate treatment strategy for each patient based on the personalized information. The availability of massive data about drugs and cell lines facilitates the possibility of proposing efficient computational models for predicting anticancer drug response. In this study, we propose ADRML, a model for Anticancer Drug Response Prediction using Manifold Learning to systematically integrate the cell line information with the drug information to make accurate predictions about drug therapeutic. The proposed model maps the drug response matrix into the lower-rank spaces that lead to obtaining new perspectives about cell lines and drugs. The drug response for a new cell line-drug pair is computed using the low-rank features. The evaluation of ADRML performance on various types of cell lines and drug information, in addition to the comparisons with previously proposed methods, shows that ADRML provides accurate and robust predictions. Further investigations about the association between drug response and pathway activity scores reveal that the predicted drug responses can shed light on the underlying drug mechanism. Also, the case studies suggest that the predictions of ADRML about novel cell line-drug pairs are validated by reliable pieces of evidence from the literature. Consequently, the evaluations verify that ADRML can be used in accurately predicting and imputing the anticancer drug response.

An Improved Anticancer Drug-Response Prediction Based on an Ensemble Method Integrating Matrix Completion and Ridge Regression

In this study, we proposed an ensemble learning method, simultaneously integrating a low-rank matrix completion model and a ridge regression model to predict anticancer drug response on cancer cell lines. The model was applied to two benchmark datasets, including the Cancer Cell Line Encyclopedia (CCLE) and the Genomics of Drug Sensitivity in Cancer (GDSC). As previous studies suggest, the dual-layer integrated cell line-drug network model was one of the best models by far and outperformed most state-of-the-art models. Thus, we performed a head-to-head comparison between the dual-layer integrated cell line-drug network model and our model by a 10-fold crossvalidation study. For the CCLE dataset, our model has a higher Pearson correlation coefficient between predicted and observed drug responses than that of the dual-layer integrated cell line-drug network model in 18 out of 23 drugs. For the GDSC dataset, our model is better in 26 out of 28 drugs in the phosphatidylinositol 3-kinase (PI3K) pathway and 26 out of 30 drugs in the extracellular signal-regulated kinase (ERK) signaling pathway, respectively. Based on the prediction results, we carried out two types of case studies, which further verified the effectiveness of the proposed model on the drug-response prediction. In addition, our model is more biologically interpretable than the compared method, since it explicitly outputs the genes involved in the prediction, which are enriched in functions, like transcription, Src homology 2/3 (SH2/3) domain, cell cycle, ATP binding, and zinc finger.

Applications of personalised signalling network models in precision oncology

As our ability to provide in-depth, patient-specific characterisation of the molecular alterations within tumours rapidly improves, it is becoming apparent that new approaches will be required to leverage the power of this data and derive the full benefit for each individual patient. Systems biology approaches are beginning to emerge within this field as a potential method of incorporating large volumes of network level data and distilling a coherent, clinically-relevant prediction of drug response. However, the initial promise of this developing field is yet to be realised. Here we argue that in order to develop these precise models of individual drug response and tailor treatment accordingly, we will need to develop mathematical models capable of capturing both the dynamic nature of drug-response signalling networks and key patient-specific information such as mutation status or expression profiles. We also review the modelling approaches commonly utilised within this field, and outline recent examples of their use in furthering the application of systems biology for a precision medicine approach to cancer treatment.

A Deep Learning Model for Cell Growth Inhibition IC50 Prediction and Its Application for Gastric Cancer Patients

Heterogeneity in intratumoral cancers leads to discrepancies in drug responsiveness, due to diverse genomics profiles. Thus, prediction of drug responsiveness is critical in precision medicine. So far, in drug responsiveness prediction, drugs’ molecular “fingerprints”, along with mutation statuses, have not been considered. Here, we constructed a 1-dimensional convolution neural network model, DeepIC50, to predict three drug responsiveness classes, based on 27,756 features including mutation statuses and various drug molecular fingerprints. As a result, DeepIC50 showed better cell viability IC50 prediction accuracy in pan-cancer cell lines over two independent cancer cell line datasets. Gastric cancer (GC) is not only one of the lethal cancer types in East Asia, but also a heterogeneous cancer type. Currently approved targeted therapies in GC are only trastuzumab and ramucirumab. Responsive GC patients for the drugs are limited, and more drugs should be developed in GC. Due to the importance of GC, we applied DeepIC50 to a real GC patient dataset. Drug responsiveness prediction in the patient dataset by DeepIC50, when compared to the other models, were comparable to responsiveness observed in GC cell lines. DeepIC50 could possibly accurately predict drug responsiveness, to new compounds, in diverse cancer cell lines, in the drug discovery process.