Öznur Taştan

The paper evaluates existing batch integration metrics in scRNA-seq and introduces sBEE, a new metric that provides more reliable assessments of batch mixing across diverse scenarios.

DeepSynBa predicts complete dose-response matrices for drug combinations, outperforming existing methods and enabling more precise and actionable drug synergy analysis.

This paper shows that using one-hot encodings for drugs and cell lines in synergy prediction models can perform as well as or better than using molecular features, suggesting models rely more on identifiers than features.

DARKIN evaluates phosphosite-dark kinase associations using zero-shot classification with protein language models, showing ESM, ProtT5-XL, and SaProt perform best.

This paper introduces PDSP, a deep learning framework that uses patient-specific single-drug response data to predict personalized drug synergy, improving accuracy by 27% in leukemia patients.

PHACTboost improves missense variant pathogenicity prediction by integrating phylogenetic trees, multiple sequence alignments, and ancestral reconstruction, outperforming existing tools including AlphaMissense and CPT-1.

FlyVISTA is a machine learning platform that identifies sleep-associated microbehaviors in Drosophila by analyzing high-resolution video data and annotating body parts.

MuDCoD detects dynamic gene communities across multiple subjects and time points in scRNA-seq data, revealing subject- and time-specific network variations.

The paper introduces three deep learning models for predicting SUMOylation sites and reports that SUMOnet-3 outperforms existing methods with 75.8% AUPR and 87% AUC.

This paper introduces Macarons, a fast algorithm for selecting complementary genetic variants to predict quantitative phenotypes, achieving similar or better performance than existing methods with significantly reduced runtime.

MatchMaker predicts drug synergy using chemical structures and gene expression, improving performance over existing methods on a large drug combination dataset.

NoRCE performs cis enrichment analysis for non-coding RNAs by examining nearby coding genes' functional annotations and integrates additional biological data to infer functional associations.

PRER improves survival prediction in 9/10 cancers by integrating protein expression with PPI networks using pairwise relative expression features.

miRCoop identifies synergistic miRNA pairs that induce strong repression on target mRNAs when combined, using kernel-based interaction tests and applies it to kidney tumor data to find 66 putative triplets.

The paper introduces an epistasis test prioritization algorithm that improves precision and reduces computational load in GWAS by selecting diverse and complementary SNPs.

The paper presents a deep learning framework to detect misannotated long non-coding RNAs by analyzing training dynamics, achieving high accuracy in distinguishing coding from non-coding sequences.

This paper identifies cross-cancer patients using a semi-supervised deep clustering approach, finding 20 patients that consistently co-cluster across different cancer types.

The paper analyzes deregulated long non-coding RNAs in cancer by examining their genomic neighbors and functional enrichment, finding cancer-specific and shared lncRNAs and functional similarities across cancer types.

PAMOGK integrates multi-omics data with pathway knowledge to cluster cancer patients, identifying subgroups with distinct survival and clinical features in kidney renal clear cell carcinoma.

DeepKinZero predicts kinase-phosphosite associations for understudied kinases using zero-shot learning, improving accuracy compared to existing methods.

This study introduces DeepSide, a deep learning framework for drug side effect prediction, showing improved performance using drug chemical structure and gene expression data.

SPADIS selects diverse SNPs for GWAS to improve predictive performance by prioritizing remote SNPs in SNP-SNP networks, outperforming SConES in regression and gene identification.

The paper identifies subtype-specific lncRNA-mediated sponge interactions in breast cancer and shows their prognostic significance.