Biology subjects

Acar, A. C.

Publications and source records attributed to Acar, A. C..

4 recordsLinked to original sources

bioRxiv2026

Lysosomal Alkalinization Selects for Metabolically Plastic, Motile Cancer Cells under Nutrient Stress.

Cancer cells exposed to nutrient deprivation activate adaptive programs to survive metabolic stress, often acquiring enhanced plasticity and motility. We have previously reported that colon cancer cell lines that survived nutrient depletion underwent partial epithelial-mesenchymal transition (pEMT), which was further exacerbated when these cells also underwent lysosomal alkalinization. Here, we have attempted to dissect the molecular mechanisms that drive the motility and shape change from cobblestone to elongated in subpopulations of cells. Using RNA-seq-based bioinformatic analyses integrated with pathway scoring, protein-protein interaction networks, probabilistic modeling and confirmatory experimental data, we have identified the coordinated activation of sublethal apoptotic signaling, fatty acid oxidation, mitochondrial ROS generation, and Ca{superscript 2}-dependent lysosomal exocytosis in the nutrient-depleted cells. Among these phenotypes, the cells undergoing starvation and lysosomal alkalinization exclusively mediated lysosomal exocytosis and cell motility. Probabilistic modeling further revealed non-linear relationships between metabolic stress signals and cell fate transitions, highlighting heterogeneous lysosomal functions as a key determinant of the altered phenotype of cells under nutrient depletion. Overall, our study has identified that aberrant lysosomal functioning in cells under nutrient depletion can specifically select for a subpopulation of cells that are highly viable, metabolically plastic and capable of motility.

cell biology↗

bioRxiv2025

Molecular Contrastive Learning with Graph Attention Network (MoCL-GAT) for Enhanced Molecular Representation

Learning the representation of molecules is crucial for drug discovery but is often hindered by the scarcity of labeled experimental data, which limits the performance of supervised machine learning models. While self-supervised learning (SSL) offers a solution by leveraging vast unlabeled chemical databases, many existing methods focus on learning from either local structural information or global molecular properties, but not both simultaneously. We introduce MoCL-GAT, a novel contrastive and transfer learning-based SSL framework that addresses this gap by simultaneously learning from two complementary objectives. It combines a local contrastive task on molecular subgraphs to capture fine-grained chemical environments with a global predictive task to learn holistic molecular descriptors. This dual-objective approach, powered by a Graph Attention Network, is designed to create more robust, versatile, and transferable molecular representations. Pre-trained on 1.9 million compounds, MoCL-GAT was fine-tuned on diverse benchmarks. It achieved state-of-the-art performance on molecular property prediction tasks, with an AUROC of 0.928 on BBBP and 0.749 on SIDER, and top-ranking RMSEs of 0.570 for ESOL and 1.818 for FreeSolv. Critically, fine-tuned models consistently and significantly outperformed models trained from scratch, confirming the value of pre-training. These results validate that MoCL-GATs dual-objective approach learns highly effective and transferable representations, enabling more accurate and data-efficient predictions for key cheminformatics challenges.

bioinformatics↗

bioRxiv2025

Development of an EMT-related exosomal miRNA signature that can predict prognosis in hepatocellular carcinoma

Chemoresistance and epithelial-mesenchymal transition (EMT) are associated with failure of cancer chemotherapy and poor survival of patients. We have previously shown that chemoresistance and stemness in hepatocellular carcinoma (HCC) cells was accompanied by the development of partial EMT (p-EMT) and identified a number of EMT-associated proteins that are released from exosomes. In this study, we aimed to identify and classify the differentially expressed (DE) exosomal miRNAs from chemoresistant HuH7 cells undergoing p-EMT. Out of the fifty-four miRNAs that were enriched in the exosomes from these cells compared to controls, thirteen were identified in the exosomes isolated from the serum of HCC patients. These miRNAs targeted genes that were associated with cell-cell junctions, extracellular matrix, cytoskeleton, transcription and signal transduction. Univariate Cox regression analysis indicated that 11/13 miRNAs were associated with either favorable (n=4) or worse (n=7) prognosis. A machine learning algorithm indicated that seven miRNAs (miR 215-5p, miR 340-5p, miR 210-3p, miR 19a-3p, miR19b-3p, miR 1266-5p and miR 25-3p) could predict worse prognosis in multiple datasets with 64-68% accuracy. A Bayesian Inference network analysis with the thirteen miRNAs and their key target proteins, along with EMT and survival as the nodes indicated that the common denominator was transcription, suggesting that the exosomal miRNAs released from cells undergoing p-EMT can mediate phenotypic changes in cells through transcriptional regulation.

cancer biology↗

bioRxiv2020

Evaluation of Methods for Protein Representation Learning: A Quantitative Analysis

Data-centric approaches have been utilized to develop predictive methods for elucidating uncharacterized aspects of proteins such as their functions, biophysical properties, subcellular locations and interactions. However, studies indicate that the performance of these methods should be further improved to effectively solve complex problems in biomedicine and biotechnology. A data representation method can be defined as an algorithm that calculates numerical feature vectors for samples in a dataset, to be later used in quantitative modelling tasks. Data representation learning methods do this by training and using a model that employs statistical and machine/deep learning algorithms. These novel methods mostly take inspiration from the data-driven language models that have yielded ground-breaking improvements in the field of natural language processing. Lately, these learned data representations have been applied to the field of protein informatics and have displayed highly promising results in terms of extracting complex traits of proteins regarding sequence-structure-function relations. In this study, we conducted a detailed investigation over protein representation learning methods, by first categorizing and explaining each approach, and then conducting benchmark analyses on; (i) inferring semantic similarities between proteins, (ii) predicting ontology-based protein functions, and (iii) classifying drug target protein families. We examine the advantages and disadvantages of each representation approach over the benchmark results. Finally, we discuss current challenges and suggest future directions. We believe the conclusions of this study will help researchers in applying machine/deep learning-based representation techniques on protein data for various types of predictive tasks. Furthermore, we hope it will demonstrate the potential of machine learning-based data representations for protein science and inspire the development of novel methods/tools to be utilized in the fields of biomedicine and biotechnology.

bioinformatics↗