A5051657734- Bioinformatics and Genomic Networks
- Cancer-related molecular mechanisms research
- Gene expression and cancer classification
- Computational Drug Discovery Methods
- RNA Research and Splicing
- Machine Learning in Bioinformatics
- Genetic Associations and Epidemiology
- Protein Structure and Dynamics
- RNA modifications and cancer
- Genetic Mapping and Diversity in Plants and Animals
- Cell Image Analysis Techniques
- MicroRNA in disease regulation
- Genomics and Phylogenetic Studies
- Machine Learning and Algorithms
- Topic Modeling
- Privacy-Preserving Technologies in Data
- Natural Language Processing Techniques
- Machine Learning in Materials Science
- vaccines and immunoinformatics approaches
- Lipid Membrane Structure and Behavior
- Genomics and Rare Diseases
- Algorithms and Data Compression
- HIV Research and Treatment
- Biomedical Text Mining and Ontologies
- Photoreceptor and optogenetics research
Sabancı Üniversitesi
2017-2025
Bilkent University
2013-2022
National Cancer Institute
2021
Dream Laboratory (United Kingdom)
2021
Framingham Heart Study
2014
National Institutes of Health
2014
Center for Information Technology
2014
National Heart Lung and Blood Institute
2014
Microsoft Research New England (United States)
2011
Carnegie Mellon University
2007-2011
Abstract Motivation: Protein–protein interactions (PPIs) are critical for virtually every biological function. Recently, researchers suggested to use supervised learning the task of classifying pairs proteins as interacting or not. However, its performance is largely restricted by availability truly (labeled). Meanwhile, there exists a considerable amount protein where an association appears between two partners, but not enough experimental evidence support it direct interaction (partially...
Drug combination therapies have been a viable strategy for the treatment of complex diseases such as cancer due to increased efficacy and reduced side effects. However, experimentally validating all possible combinations synergistic interaction even with high-throughout screens is intractable vast combinatorial search space. Computational techniques can reduce number be evaluated by prioritizing promising candidates. We present MatchMaker that predicts drug synergy scores using chemical...
Background— Cardiovascular disease (CVD) reflects a highly coordinated complex of traits. Although genome-wide association studies have reported numerous single nucleotide polymorphisms (SNPs) to be associated with CVD, the role most these variants in processes remains unknown. Methods and Results— We built CVD network using 1512 SNPs 21 traits (at P ≤5×10 −8 ) cross-linked different by virtue their shared SNP associations. then explored whole blood gene expression relation 5257 participants...
Long non-coding RNAs (lncRNAs) can indirectly regulate mRNAs expression levels by sequestering microRNAs (miRNAs), and act as competing endogenous (ceRNAs) or sponges. Previous studies identified lncRNA-mediated sponge interactions in various cancers including the breast cancer. However, cancer subtypes are quite distinct terms of their molecular profiles; therefore, ceRNAs expected to be subtype-specific well. To find ceRNA subtypes, we develop an integrative approach. We conduct partial...
Drug failures due to unforeseen adverse effects at clinical trials pose health risks for the participants and lead substantial financial losses. Side effect prediction algorithms have potential guide drug design process. LINCS L1000 dataset provides a vast resource of cell line gene expression data perturbed by different drugs creates knowledge base context specific features. The state-of-the-art approach that aims using information relies on only high-quality experiments in discards large...
Motivation: Many cancer monotherapies demonstrate limited clinical efficacy, making combination therapies a relevant treatment strategy. The extensive number of potential drug combinations and context-specific response profiles complicates the prediction responses. Existing computational models are typically trained to predict single aggregated synergy score, which summarises responses across different dosage combinations, such as Bliss or Loewe scores. This oversimplification drug-response...
There is great interest in using genetically tractable organisms such as Drosophila to gain insights into the regulation and function of sleep. However, sleep phenotyping has largely relied on simple measures locomotor inactivity. Here, we present FlyVISTA, a machine learning platform perform deep flies. This comprises high-resolution closed-loop video imaging system, coupled with network annotate 35 body parts, computational pipeline extract behaviors from high-dimensional data. FlyVISTA...
As single-cell genomics experiments increase in complexity and scale, the need to integrate multiple datasets has grown. Such integration enhances cellular feature identification by leveraging larger data volumes. However, batch effects-technical variations arising from differences labs, times, or protocols-pose a significant challenge. Despite numerous proposed correction methods, many still have limitations, such as outputting only dimension-reduced data, relying on computationally...
Abstract Motivation Protein phosphorylation is a key regulator of protein function in signal transduction pathways. Kinases are the enzymes that catalyze other proteins target-specific manner. The dysregulation associated with many diseases including cancer. Although advances phosphoproteomics enable identification phosphosites at proteome level, most phosphoproteome still dark: more than 95% reported human have no known kinases. Determining which kinase responsible for phosphorylating site...
Individuals (and their family members) share (partial) genomic data on public platforms. However, using special characteristics of data, background knowledge that can be obtained from the Web, and relationship between individuals, it is possible to infer hidden parts shared unshared) genomes. Existing work in this field considers simple correlations genome (as well as Mendel's law partial genomes a victim his members). In paper, we improve existing inference attacks privacy. We mainly...
Abstract Background The relationships between tonsillar immune responses, and viral infection allergy are incompletely known. Objective To study intratonsillar/nasopharyngeal virus detections in vivo expressions of T ‐cell‐ innate response‐specific cytokines, transcription factors, type I/II/III interferons human tonsils. Methods Palatine tonsil samples were obtained from 143 elective tonsillectomy patients. Adenovirus, bocavirus‐1, coronavirus, enteroviruses, influenza virus,...
Abstract Drug failures due to unforeseen adverse effects at clinical trials pose health risks for the participants and lead substantial financial losses. Side effect prediction algorithms have potential guide drug design process. LINCS L1000 dataset provides a vast resource of cell line gene expression data perturbed by different drugs creates knowledge base context specific features. The state-of-the-art approach that aims using information relies on only high-quality experiments in...
Abstract Motivation Accurate classification of patients into molecular subgroups is critical for the development effective therapeutics and deciphering what drives these to cancer. The availability multiomics data catalogs large cohorts cancer provides multiple views biology tumors with unprecedented resolution. Results We develop Pathway-based MultiOmic Graph Kernel clustering (PAMOGK) that integrates patient existing biological knowledge on pathways. a novel graph kernel evaluates...
Abstract HIV-1 Vpr, a nonstructural viral protein associated with virus particles, has positive role in the efficient transport of PIC into nucleus non-dividing target cells and enhances replication primary T cells. Vpr is 96 amino acid structure by NMR shows three helical domains. been shown to exist as dimers higher order oligomers. Considering multifunctional nature contribution distinct domains dimer/oligomer relevance this feature its functions are not clear. To address this, we have...
Viruses depend on their hosts at every stage of life cycles and must therefore communicate with them via Protein-Protein Interactions (PPIs). To investigate the mechanisms communication by different viruses, we overlay reported pairwise human-virus PPIs human signalling pathways. Of 671 pathways obtained from NCI Reactome databases, 355 are potentially targeted least one virus. The majority linked to more than We find evidence supporting hypothesis that viruses often interact proteins...
Abstract Drug combination therapies have been a viable strategy for the treatment of complex diseases such as cancer due to increased efficacy and reduced side effects. However, experimentally validating all possible combinations synergistic interaction even with high-throughout screens is intractable vast combinatorial search space. Computational techniques can reduce number be evaluated by prioritizing promising candidates. We present MatchMaker that predicts drug synergy scores using...
Developing a better mechanistic understanding of membrane protein folding is urgently needed because the discovery an increasing number human diseases, where instability and misfolding involved. Towards this goal, we investigated stability 7-transmembrane (TM) helical bundles by computational methods. We compared results three different algorithms for predicting changes in proteins against experimental mutation dataset obtained bacteriorhodopsin (BR) mammalian rhodopsin find that 61.6% 70.6%...
Abstract Background Human immunodeficiency virus-1 (HIV-1) has a minimal genome of only 9 genes, which encode 15 proteins. HIV-1 thus depends on the human host for virtually every aspect its life cycle. The universal language communication in biological systems, including between pathogen and host, is via signal transduction pathways. fundamental units these pathways are protein interactions. Understanding functional significance HIV-1, interactions requires viewing them context Results...
The last decade has proven that amyotrophic lateral sclerosis (ALS) is clinically and genetically heterogeneous, the genetic component in sporadic cases might be stronger than expected. This study investigates 1,200 patients to revisit ALS ethnically heterogeneous yet inbred Turkish population. Familial (fALS) accounts for 20% of our cases. rates consanguinity are 30% fALS 23% (sALS). Major genes explained disease cause only 35% fALS, as compared with ~70% Europe North America. Whole exome...
Machine learning could be used to predict the likelihood of diagnosis and severity illness. Lack COVID-19 patient data has hindered science community in developing models aid response pandemic.
Abstract Motivation Genomic studies identify genomic loci representing genetic variations, transcription factor (TF) occupancy, or histone modification through next generation sequencing (NGS) technologies. Interpreting these requires evaluating them with known and epigenomic annotations. Results We present GLANET as a comprehensive annotation enrichment analysis tool which implements sampling-based test that accounts for GC content and/or mappability biases, jointly separately. annotates...
Rapid and low cost sequencing of genomes enabled widespread use genomic data in research studies personalized customer applications, where is shared public databases. Although the identities participants are anonymized these databases, sensitive information about individuals can still be inferred. One such kinship.We define two routes kinship privacy leak propose a technique to protect against risks while maximizing utility data. The method involves systematic identification minimal portions...