Biology subjects

Bozdag, S.

Publications and source records attributed to Bozdag, S..

23 records · Page 2Linked to original sources

bioRxiv2020

CTDPathSim: Cell line-tumor deconvoluted pathway-based similarity in the context of precision medicine in cancer

In cancer research and drug development, human tumor-derived cell lines are used as popular model for cancer patients to evaluate the biological functions of genes, drug efficacy, side-effects, and drug metabolism. Using these cell lines, the functional relationship between genes and drug response and prediction of drug response based on genomic and chemical features have been studied. Knowing the drug response on the real patients, however, is a more important and challenging task. To tackle this challenge, some studies integrate data from primary tumors and cancer cell lines to find associations between cell lines and tumors. These studies, however, do not integrate multi-omics datasets to their full extent. Also, several studies rely on a genome-wide correlation-based approach between cell lines and bulk tumor samples without considering the heterogeneous cell population in bulk tumors. To address these gaps, we developed a computational pipeline, CTDPathSim, a pathway activity-based approach to compute similarity between primary tumor samples and cell lines at genetic, genomic, and epigenetic levels integrating multi-omics datasets. We utilized a deconvolution method to get cell type-specific DNA methylation and gene expression profiles and computed deconvoluted methylation and expression profiles of tumor samples. We assessed CTDPathSim by applying on breast and ovarian cancer data in The Cancer Genome Atlas (TCGA) and cancer cell lines data in the Cancer Cell Line Encyclopedia (CCLE) databases. Our results showed that highly similar sample-cell line pairs have similar drug response compared to lowly similar pairs in several FDA-approved cancer drugs, such as Paclitaxel, Vinorelbine and Mitomycin-c. CTDPathSim outperformed state-of-the-art methods in recapitulating the known drug responses between samples and cell lines. Also, CTDPathSim selected higher number of significant cell lines belonging to the same cancer types than other methods. Furthermore, our aligned cell lines to samples were found to be clinical biomarkers for patients survival whereas unaligned cell lines were not. Our method could guide the selection of appropriate cell lines to be more intently serve as proxy of patient tumors and could direct the pre-clinical translation of drug testing into clinical platform towards the personalized therapies. Furthermore, this study could guide the new uses for old drugs and benefits the development of new drugs in cancer treatments. CCS CONCEPTSO_LIComputational biology C_LIO_LIGenomics C_LIO_LISystems biology C_LIO_LIBioinformatics C_LIO_LIGenetics C_LI ACM Reference formatBanabithi Bose, Serdar Bozdag. 2020. CTDPathSim: Cell line-tumor deconvoluted pathway-based similarity in the context of precision medicine in cancer.

bioinformatics↗

bioRxiv2020

An Autoencoder and Artificial Neural Network-based Method to Estimate Parity Status of Wild Mosquitoes from Near-infrared Spectra

BackgroundAfter mating, female mosquitoes need animal blood to develop their eggs. In the process of acquiring blood, they may acquire pathogens, which may cause different diseases to humans such as malaria, zika, dengue, and chikungunya. Therefore, knowing the parity status of mosquitoes is useful in control and evaluation of infectious diseases transmitted by mosquitoes, where parous mosquitoes are assumed to be potentially infectious. Ovary dissections, which currently are used to determine the parity status of mosquitoes, are very tedious and limited to very few experts. An alternative to ovary dissections is near-infrared spectroscopy (NIRS), which can estimate the age in days and the infectious state of laboratory and semi-field reared mosquitoes with accuracies between 80 and 99%. No study has tested the accuracy of NIRS for estimating the parity status of wild mosquitoes. Methods and resultsIn this study, we train artificial neural network (ANN) models on NIR spectra to estimate the parity status of wild mosquitoes. We use four different datasets: An. arabiensis collected from Minepa, Tanzania (Minepa-ARA); An. gambiae collected from Muleba, Tanzania (Muleba-GA); An. gambiae collected from Burkina Faso (Burkina-GA); and An.gambiae from Muleba and Burkina Faso combined (Muleba-Burkina-GA). We train ANN models on datasets with spectra preprocessed according to previous protocols. We then use autoencoders to reduce the spectra feature dimensions from 1851 to 10 and re-train ANN models. Before the autoencoder was applied, ANN models estimated parity status of mosquitoes in Minepa-ARA, Muleba-GA, Burkina-GA and Muleba-Burkina-GA with out-of-sample accuracies of 81.9 {+/-} 2.8% (N=927), 68.7 {+/-} 4.8% (N=140), 80.3 {+/-} 2.0% (N=158), and 75.7 {+/-} 2.5% (N=298), respectively. With the autoencoder, ANN models tested on out-of-sample data achieved 97.1 {+/-} 2.2%, (N=927), 89.8 {+/-} 1.7% (N=140), 93.3 {+/-} 1.2% (N=158), and 92.7 {+/-} 1.8% (N=298) accuracies for Minepa-ARA, Muleba-GA, Burkina-GA, and Muleba-Burkina-GA, respectively. ConclusionThese results show that a combination of an autoencoder and an ANN trained on NIR spectra to estimate parity status of wild mosquitoes yields models that can be used as an alternative tool to estimate parity status of wild mosquitoes, especially since NIRS is a high-throughput, reagent-free, and simple-to-use technique compared to ovary dissections.

bioinformatics↗

bioRxiv2019

miRDriver: A Tool to Infer Copy Number Derived miRNA-Gene Networks in Cancer

Copy number aberration events such as amplifications and deletions in chromosomal regions are prevalent in cancer patients. Frequently aberrated copy number regions include regulators such as microRNAs (miRNAs), which regulate downstream target genes that involve in the important biological processes in tumorigenesis and proliferation. Many previous studies explored the miRNA-gene interaction networks but copy number-derived miRNA regulations are limited. Identifying copy number-derived miRNA-target gene regulatory interactions in cancer could shed some light on biological mechanisms in tumor initiation and progression. In the present study, we developed a computational pipeline, called miRDriver which is based on the hypothesis that copy number data from cancer patients can be utilized to discover driver miRNAs of cancer. miRDriver integrates copy number aberration, DNA methylation, gene and miRNA expression datasets to compute copy number-derived miRNA-gene interactions in cancer. We tested miRDriver on breast cancer and ovarian cancer data from the Cancer Genome Atlas (TCGA) database. miRDriver discovered some of the known miRNAs, such as miR-125b, mir-320d, let-7g, and miR-21, which are known to be in copy number aberrated regions in breast cancer. We also discovered some potentially novel miRNA-gene interactions. Also, several miRNAs such as miR-127, miR-139 and let-7b were found to be associated with tumor survival and progression based on Cox proportional hazard model. We compared the enrichment of known miRNA-gene interactions computed by miRDriver with the enrichment of interactions computed by the state-of-the-art methods and miRDriver outperformed all the other methods.\n\nCCS CONCEPTSO_LIBioinformatics\nC_LIO_LIComputational Genomics\nC_LIO_LIBiological Networks\nC_LI

bioinformatics↗

bioRxiv2019

PhenoGeneRanker: A Tool for Gene Prioritization Using Complete Multiplex Heterogeneous Networks

Uncovering genotype-phenotype relationships is a fundamental challenge in genomics. Gene prioritization is an important step for this endeavor to make a short manageable list from a list of thousands of genes coming from high-throughput studies. Network propagation methods are promising and state of the art methods for gene prioritization based on the premise that functionally-related genes tend to be close to each other in the biological networks.\n\nIn this study, we present PhenoGeneRanker, an improved version of a recently developed network propagation method called Random Walk with Restart on Multiplex Heterogeneous Networks (RWR-MH). PhenoGeneRanker allows multi-layer gene and disease networks. It also calculates empirical p-values of gene ranking using random stratified sampling of genes based on their connectivity degree in the network.\n\nWe ran PhenoGeneRanker using multi-omics datasets of rice to effectively prioritize the cold tolerance-related genes. We observed that top genes selected by PhenoGeneRanker were enriched in cold tolerance-related Gene Ontology (GO) terms whereas bottom ranked genes were enriched in general GO terms only. We also observed that top-ranked genes exhibited significant p-values suggesting that their rankings were independent of their degree in the network.\n\nCCS CONCEPTS* Bioinformatics * Biological networks * System biology * Computational genomics\n\nAvailability and implementationThe source code is available on GitHub at https://github.com/bozdaglab/PhenoGeneRanker under Creative Commons Attribution 4.0 license\n\nContactcdursun@mcw.edu or serdar.bozdag@marquette.edu

bioinformatics↗

bioRxiv2019

HAVE IT, KNOW IT, BUT DO NOT SHOW IT: EXAMINING PHYSIOLOGICAL AROUSAL, ANXIETY, AND FACIAL EXPRESSIONS OVER THE COURSE OF A SOCIAL SKILLS INTERVENTION FOR AUTISTIC ADOLESCENTS

Facial expressions provide a nonverbal mechanism for social communication, a core challenge for autistic people. Little is known regarding the association between arousal, self-report of anxiety, and facial expressions among autistic adolescents. Therefore, this study investigated session-by-session facial expressions, self-report of anxiety, and physiological arousal via Electrodermal Activity (EDA), of 12 autistic male adolescents in a didactic social skills intervention setting. The goals of this study were threefold: 1) identify physiological arousal levels (\"have-it\"), 2) examine if autistic adolescents facial expressions indicated arousal (\"show-it\"), and 3) determine whether autistic adolescents were self-aware of their anxiety (\"know-it\"). Our results showed that autistic adolescents self-rated anxiety was significantly associated with peaks in EDA. Both machine learning algorithms and human participant-based methods, however, had low accuracy in predicting autistic adolescents arousal state from facial expressions, suggesting that autistic adolescents facial expressions did not coincide with their arousal. Implications for understanding social communication difficulties among autistic adolescents, as well as future targets for intervention, are discussed. This project is registered with ClinicalTrials.gov, Identifier: NCT02680015.

bioinformatics↗