Weijia Cai

DNA foundation models trained on large-scale genomic and epigenetic datasets have shown promise for regulatory variant interpretation, yet their application to tissue-specific contexts remain limited. Here, we present a transfer learning (TL) framework to adapt Enformer, a deep neural network trained on 5,313 multi-omics tracks, to breast and prostate cancer using 275 and 357 tissue-specific transcription factor (TF) ChIP–seq tracks, respectively. We computed tissue-specific cis-regulatory activity (tCRA) scores for millions of single-nucleotide variants (SNVs) in genome-wide association study (GWAS) datasets and prioritized high-impact SNV subsets (1M, 1.5M, and 2M). These TL-prioritized variants demonstrated consistently greater enrichment in tissue-specific enhancers, cancer GWAS risk variants, and ClinVar pathogenic variants compared to the original Enformer model. Transcriptome-wide association studies (TWAS) using TL-based SNVs identified more cancer-relevant genes, many of which exhibited functional essentiality (DepMap), therapeutic tractability (drug databases), and disease relevance (DisGeNET). Notably, TL models outperformed the base model in identifying genes enriched for drug targets and clinically relevant disease associations. Our results show that TL-derived tCRA scores enhance regulatory variant prioritization and improve susceptibility gene discovery in a tissue-specific manner. Our study provides a generalizable framework for tailoring foundation models to disease-relevant contexts, with implications for variant interpretation, therapeutic target discovery, and precision medicine.

Supplementary Figure from A Genome-Wide Screen Identifies PDPK1 as a Target to Enhance the Efficacy of MEK1/2 Inhibitors in NRAS Mutant Melanoma

Supplementary Table from A Genome-Wide Screen Identifies PDPK1 as a Target to Enhance the Efficacy of MEK1/2 Inhibitors in NRAS Mutant Melanoma

Supplementary Figure from A Genome-Wide Screen Identifies PDPK1 as a Target to Enhance the Efficacy of MEK1/2 Inhibitors in NRAS Mutant Melanoma

Supplementary Table from A Genome-Wide Screen Identifies PDPK1 as a Target to Enhance the Efficacy of MEK1/2 Inhibitors in NRAS Mutant Melanoma

Supplementary Table from A Genome-Wide Screen Identifies PDPK1 as a Target to Enhance the Efficacy of MEK1/2 Inhibitors in NRAS Mutant Melanoma

Melanomas frequently harbor activating NRAS mutations. However, limited advance has been made in developing targeted therapy options for patients with NRAS mutant melanoma. MEK inhibitors (MEKi) show modest efficacy in the clinic and their actions need to be optimized. In this study, we performed a genome-wide CRISPR-Cas9–based screen and demonstrated that loss of phosphoinositide-dependent kinase-1 (PDPK1) enhances the efficacy of MEKi. The synergistic effects of PDPK1 loss and MEKi was validated in NRAS mutant melanoma cell lines using pharmacologic and molecular approaches. Combined PDPK1 inhibitors (PDPK1i) with MEKi suppressed NRAS mutant xenograft growth and induced gasdermin E–associated pyroptosis. In an immune-competent allograft model, PDPK1i+MEKi increased the ratio of intratumoral CD8+ T cells, delayed tumor growth, and prolonged survival; the combination treatment was less effective against tumors in immune-deficient mice. These data suggest PDPK1i+MEKi as an efficient immunostimulatory strategy against NRAS mutant melanoma.Significance:Targeting PDPK1 stimulates antitumor immunity and sensitizes NRAS mutant melanoma to MEK inhibition, providing rationale for the clinical development of a combinatorial approach for treating patients with melanoma.

Supplementary Figure from A Genome-Wide Screen Identifies PDPK1 as a Target to Enhance the Efficacy of MEK1/2 Inhibitors in NRAS Mutant Melanoma

Supplementary Figure from A Genome-Wide Screen Identifies PDPK1 as a Target to Enhance the Efficacy of MEK1/2 Inhibitors in NRAS Mutant Melanoma

Supplementary Figure from A Genome-Wide Screen Identifies PDPK1 as a Target to Enhance the Efficacy of MEK1/2 Inhibitors in NRAS Mutant Melanoma