Identifying Lethal Dependencies with HUGE Predictive Power

Prognostic and chemotherapeutic response prediction by proliferation essential gene signature: Investigating POLE2 in bladder cancer progression and cisplatin resistance

Background: Bladder cancer (BLCA) is the most common genitourinary malignancy. Proliferation essential genes (PEGs) are crucial to the survival of cancer cells. This study aimed to build a PEG signature to predict BLCA prognosis and treatment efficacy. Methods: BLCA PEGs and differentially expressed PEGs were identified using DepMap and TCGA-BLCA datasets, respectively. Based on the prognostic analysis of the differentially expressed PEGs, a PEG model was constructed. Subsequently, we analyzed the relationship between the PEG signature and prognosis of BLCA patients as well as their response to chemotherapy. Finally, we performed random forest analysis to target and functional experiments to validate the most significant PEG which is associated with BLCA progression. CCK-8, invasion, migration, and chemosensitivity assays were performed to assess effects of gene knockdown on BLCA cell proliferation, invasion and migration abilities, and cisplatin chemosensitivity. Results: We screened 10 prognostic PEGs from 201 differentially expressed PEGs and used them to construct a PEG signature model. Patients with high PEG signature score (PEGs-high) exhibited worse OS and lower sensitivity to chemotherapy than those with PEGs-low. We also found significant correlations between the PEG score and previously defined BLCA molecular subtypes. This suggests that the PEG score may effectively predict the molecular subtypes which have distinct clinical outcomes. Random forest analysis revealed that POLE2 (DNA polymerase epsilon subunit 2) was the most significant PEG differentiating BLCA tissue and normal tissue. Bioinformatic analysis and an immunohistochemistry staining assay confirmed that POLE2 was significantly up-regulated in tumor tissues and was associated with poor survival in BLCA patients. Moreover, POLE2 knockdown inhibited the ability of cell clone formation, proliferation, invasion, immigration and IC50 of cisplatin. Conclusion: The PEG signature acts as a potential predictor for prognosis and chemotherapy response in BLCA patients. POLE2 is a key PEG and plays a remarkable role in promoting the malignant progression and cisplatin resistance of BLCA.

CRISPR single base-editing: in silico predictions to variant clonal cell lines

Engineering single base edits using CRISPR technology including specific deaminases and single-guide RNA (sgRNA) is a rapidly evolving field. Different types of base edits can be constructed, with cytidine base editors (CBEs) facilitating transition of C-to-T variants, adenine base editors (ABEs) enabling transition of A-to-G variants, C-to-G transversion base editors (CGBEs) and recently adenine transversion editors (AYBE) that create A-to-C and A-to-T variants. The base-editing machine learning algorithm BE-Hive predicts which sgRNA and base editor combinations have the strongest likelihood of achieving desired base edits. We have used BE-Hive and TP53 mutation data from The Cancer Genome Atlas (TCGA) ovarian cancer cohort to predict which mutations can be engineered, or reverted to wild-type (WT) sequence, using CBEs, ABEs or CGBEs. We have developed and automated a ranking system to assist in selecting optimally designed sgRNA that considers the presence of a suitable protospacer adjacent motif (PAM), the frequency of predicted bystander edits, editing efficiency and target base change. We have generated single constructs containing ABE or CBE editing machinery, an sgRNA cloning backbone and an enhanced green fluorescent protein tag (EGFP), removing the need for co-transfection of multiple plasmids. We have tested our ranking system and new plasmid constructs to engineer the p53 mutants Y220C, R282W and R248Q into WT p53 cells and shown that these mutants cannot activate four p53 target genes, mimicking the behaviour of endogenous p53 mutations. This field will continue to rapidly progress, requiring new strategies such as we propose to ensure desired base-editing outcomes.

Synthetic lethality in large-scale integrated metabolic and regulatory network models of human cells

Synthetic lethality (SL) is a promising concept in cancer research. A wide array of computational tools has been developed to predict and exploit synthetic lethality for the identification of tumour-specific vulnerabilities. Previously, we introduced the concept of genetic Minimal Cut Sets (gMCSs), a theoretical approach to SL developed for genome-scale metabolic networks. The major challenge in our gMCS framework is to go beyond metabolic networks and extend existing algorithms to more complex protein-protein interactions. In this article, we take a step further and incorporate linear regulatory pathways into our gMCS approach. Extensive algorithmic modifications to compute gMCSs in integrated metabolic and regulatory models are presented in detail. Our extended approach is applied to calculate gMCSs in integrated models of human cells. In particular, we integrate the most recent genome-scale metabolic network, Human1, with 3 different regulatory network databases: Omnipath, Dorothea and TRRUST. Based on the computed gMCSs and transcriptomic data, we discovered new essential genes and their associated synthetic lethal for different cancer cell lines. The performance of the different integrated models is assessed with available large-scale in-vitro gene silencing data. Finally, we discuss the most relevant gene essentiality predictions based on published literature in cancer research.

Explainable artificial intelligence for precision medicine in acute myeloid leukemia

Artificial intelligence (AI) can unveil novel personalized treatments based on drug screening and whole-exome sequencing experiments (WES). However, the concept of “black box” in AI limits the potential of this approach to be translated into the clinical practice. In contrast, explainable AI (XAI) focuses on making AI results understandable to humans. Here, we present a novel XAI method -called multi-dimensional module optimization (MOM)- that associates drug screening with genetic events, while guaranteeing that predictions are interpretable and robust. We applied MOM to an acute myeloid leukemia (AML) cohort of 319 ex-vivo tumor samples with 122 screened drugs and WES. MOM returned a therapeutic strategy based on the FLT3, CBFβ-MYH11, and NRAS status, which predicted AML patient response to Quizartinib, Trametinib, Selumetinib, and Crizotinib. We successfully validated the results in three different large-scale screening experiments. We believe that XAI will help healthcare providers and drug regulators better understand AI medical decisions.