Leggate, D. | bioRxiv Science

bioRxiv2024

Signaling pathway evaluation of leading ATRi, PARPi and CDK7i cancer compounds targeting the DNA Damage Response using Causal Inference

IntroductionThere are many cancer drugs in development which target the DNA damage response (DDR), following early successes of drugs such as olaparib. However, various challenges to the success of these inhibitors exist, including the emergence of resistance, the identification of appropriate biomarkers to identify patients who will respond to treatment, as well as the identification of combination therapies that improve efficacy without a concomitant increase in toxicity. While the identification of biomarkers of resistance could aid in overcoming these challenges, current methods mostly generate lists of potential genes, proteins that display changes in cancer patients, without exposing the underlying, and often critical, mechanisms of resistance. MethodsWe have developed the Adaptable Large-Scale Causal Analysis (ALaSCA) software platform, which applies Pearlian Causal Inference (PCI) techniques to specifically transcriptomic, proteomic and phenotypic multi-omics data. ALaSCA quantifies the causal contributions of different biological pathways to an outcome such as responsiveness to treatment. The strength of applying PCI to biological pathways lies in quantifying the causal contributions of targets, through their related pathways, to drug sensitivity - as opposed to merely enriching or grouping lists of genes into pathways. We applied ALaSCA to transcriptomic data for several different compounds related to three known inhibitor types that target DDR proteins: an ATR, a CDK7, and several PARP inhibitors. Our aim was to use causal methods to evaluate biological signaling pathways to identify resistance mechanisms that can be used for patient stratification and development of combination therapies in breast, ovarian and non-small cell lung cancer (NSCLC). Key findingsWe observed that niraparib seems to have a different resistance mechanism than other PARPi inhibitors in breast and NSCLC, which is driven by CDK1 as opposed to base excision or nucleotide excision repair. Additionally, CDK7 appears to be a significant driver of PARP inhibitor resistance, especially for niraparib, through predominantly the G2/M cell cycle phase and to a lesser extent nucleotide excision repair in breast and ovarian cancer, but not in NSCLC. Lastly, we identified that genes from the homologous recombination pathway drive resistance to AZD6738, an ATR inhibitor, in breast cancer, and that AZD6738 and irinotecan have differing resistance mechanisms in ovarian cancer, indicating the potential of combining these treatments. Next stepsOur findings demonstrate the potential of ALaSCA to generate interesting insights for treatments when applied to public data and well-known inhibitors. Partnership with industry drug discovery groups using proprietary data to rerun the above evaluations will further refine and confirm these findings.

bioinformatics↗

bioRxiv2021

Predicting genes associated with RNA methylation pathways using machine learning

RNA methylation plays an important role in functional regulation of RNAs, and has thus attracted an increasing interest in biology and drug discovery. Here, we collected and collated transcriptomic, proteomic, structural and physical interaction data from the Harmonizome database, and applied supervised machine learning to predict novel genes associated with RNA methylation pathways in human. We selected five types of classifiers, which we trained and evaluated using cross-validation on multiple training sets. The best models reached 88% accuracy based on cross-validation, and an average 91% accuracy on the test set. Using protein-protein interaction data, we propose six molecular sub-networks linking model predictions to previously known RNA methylation genes, with roles in mRNA methylation, tRNA processing, rRNA processing, but also protein and chromatin modifications. Our study exemplifies how access to large omics datasets joined by machine learning methods can be used to predict gene function.

bioinformatics↗

bioRxiv2020

Genome-wide investigation of gene-cancer associations for the prediction of novel therapeutic targets in oncology

A major cause of failed drug discovery programs is suboptimal target selection, resulting in the development of drug candidates that are potent inhibitors, but ineffective at treating the disease. In the genomics era, the availability of large biomedical datasets with genome-wide readouts has the potential to transform target selection and validation. In this study we investigate how computational intelligence methods can be applied to predict novel therapeutic targets in oncology. We compared different machine learning classifiers applied to the task of drug target classification for nine different human cancer types. For each cancer type, a set of "known" target genes was obtained and equally-sized sets of "non-targets" were sampled multiple times from the human protein-coding genes. Models were trained on mutation, gene expression (TCGA), and gene essentiality (DepMap) data. In addition, we generated a numerical embedding of the interaction network of protein-coding genes using deep network representation learning and included the results in the modeling. We assessed feature importance using a random forests classifier and performed feature selection based on measuring permutation importance against a null distribution. Our best models achieved good generalization performance based on the AUROC metric. With the best model for each cancer type, we ran predictions on more than 15,000 protein-coding genes to identify potential novel targets. Our results indicate that this approach may be useful to inform early stages of the drug discovery pipeline.

bioinformatics↗