{"doi":"10.3390/genes16101223","title":"Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants","abstract":"BACKGROUND/OBJECTIVE: Genome-wide association studies (GWAS) have linked many noncoding variants to complex traits and diseases, but distinguishing as-sociation from causation remains difficult. Deep learning models-particularly CNN- and Transformer-based architectures-are widely used for this task, yet comparisons are hindered by inconsistent benchmarks and evaluation practices. We aimed to establish a standardized assessment of leading models for predicting variant effects in enhancers and for prioritizing putative causal SNPs. METHODS: We evaluated state-of-the-art deep learning models under consistent training and evaluation conditions on nine datasets derived from MPRA, raQTL, and eQTL ex-periments. These datasets profile the regulatory impact of 54,859 single-nucleotide polymorphisms (SNPs) across four human cell lines. Performance was compared for two related tasks: predicting the direction and magnitude of regulatory impact in enhancers and identifying likely causal SNPs within linkage disequilibrium (LD) blocks. We addi-tionally assessed the effect of fine-tuning on Transformer-based models and the impact of certainty in experimental results. RESULTS: CNN models such as TREDNet and SEI performed best for predicting the reg-ulatory impact of SNPs in enhancers. Hybrid CNN-Transformer models (e.g., Borzoi) performed best for causal variant prioritization within LD blocks. Fine-tuning benefits Transformers but remains insufficient to close the performance gap. CONCLUSIONS: Under a unified benchmark, CNN architectures are most reliable for esti-mating enhancer regulatory effects of SNPs, while hybrid CNN-Transformer models are superior for causal SNP identification within LD. These comparisons help guide model selection for variant-effect prediction in noncoding regions.","journal":"Genes","year":2025,"id":527218,"datarank":0.20794415416798362,"base_score":1.3862943611198906,"endowment":1.3862943611198906,"self_citation_contribution":0.20794415416798362,"citation_network_contribution":0.0,"self_endowment_contribution":0.20794415416798362,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":3,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9308,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2025-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":1404538,"name":"Kathryn Borkowski","orcid":null,"position":1,"is_corresponding":false},{"id":54330,"name":"Ivan Ovcharenko","orcid":"0000-0002-9730-7732","position":2,"is_corresponding":false},{"id":1072811,"name":"Gaetano Manzo","orcid":"0000-0002-1682-6865","position":0,"is_corresponding":true}],"reference_count":35,"raw_metadata":{"citation_network_status":"fetched"},"created_at":"2026-07-19T02:50:39.280101Z","pmid":"41153440","pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}