Biology subjects

Gursoy, A.

Publications and source records attributed to Gursoy, A..

13 recordsLinked to original sources

bioRxiv2025

Network-Based Analysis of Human Astrocytes Links Aging to Neurodegenerative and Cardiovascular Diseases

Astrocytes are central to brain homeostasis, supporting neuronal metabolism, synaptic activity, and the blood-brain barrier. With aging, these glial cells undergo molecular and functional changes that weaken support functions and promote neuroinflammation, contributing to neurodegeneration. Yet the systems-level mechanisms of astrocytic aging remain poorly defined in human models. Because aging also heightens risk for cardiovascular disease, cognitive impairment, type 2 diabetes, and systemic inflammation, clarifying shared astrocytic pathways is critical for understanding brain-body crosstalk. Using an in vitro human astrocyte model exposed to sublethal oxidative stress (10 {micro}M H2O2), we profiled transcriptomic changes and identified differentially expressed genes across antioxidant defences, proteostasis, transcriptional regulation, vesicular trafficking, and inflammatory signalling. We then performed seven network-prioritization analyses on a curated human protein-protein interactome: one seeded with the astrocyte H2O2-responsive genes and six with phenotype-associated gene sets (Alzheimers disease, cardiovascular disease, cognitive impairment, type 2 diabetes, oxidative stress, and inflammation). Intersecting the top 5% scoring genes from each run yielded a 127-gene core shared across all seven, enriched for proteostasis, DNA repair, mitochondrial regulation, and telomere and nuclear envelope maintenance. Structure-guided analyses highlighted vulnerable interfaces, including lamin A/C-lamin B1, -actinin-filamins, 14-3-3 dimers, and aminoacyl-tRNA synthetase assemblies, where pathogenic variants are predicted to destabilize or aberrantly stabilize protein interactions. Structure-based interface predictions also highlight potential interactions between APP-VCP/p97 and p53-14-3-3{zeta} that link proteostasis and stress signalling. Together, these findings define a conserved astrocytic vulnerability network that may couple neurodegeneration with cardiovascular disease and nominate structurally testable targets for biomarkers and interventions.

bioinformatics↗

bioRxiv2024

DeepAllo: Allosteric Site Prediction using Protein Language Model (pLM) with Multitask Learning

Allostery, the process by which binding at one site perturbs a distant site, is being rendered as a key focus in the field of drug development with its substantial impact on protein function. The identification of allosteric pockets (sites) is a challenging task and several techniques have been developed, including Machine Learning (ML) to predict allosteric pockets that utilize both static and pocket features. Our work, DeepAllo, is the first study that combines fine-tuned protein language model (pLM) with FPocket features and shows an increase in prediction performance of allosteric sites over previous studies. The pLM model was fine-tuned on Allosteric Dataset (ASD) in Multitask Learning (MTL) setting and was further used as a feature extractor to train XGBoost and AutoML models. The best model predicts allosteric pockets with 89.66% F1 score and 90.5% of allosteric pockets in the top 3 positions, outperforming previous results. A case study has been performed on proteins with known allosteric pockets, which shows the proof of our approach. Moreover, an effort was made to explain the pLM by visualizing its attention mechanism among allosteric and non-allosteric residues.

bioinformatics↗

bioRxiv2024

An Inflammation Centered Perspective to the Mechanisms and Interactions Related to Vascular Cognitive Impairment

A major health burden for the elderly, vascular cognitive impairment (VCI) is a disease that combines cognitive (CD) and cardiovascular (CVD) components. The molecular mechanisms underlying this disease are poorly understood, and our work attempts to bridge this knowledge gap by building protein-protein interaction (PPI) networks of CD and CVD. Our earlier research not only showed how well these two primary components work together, but also hinted at the potential role of inflammation in the development of VCI. For this reason, we decided to examine the relationship between inflammation and VCI in further detail.We identified the top three most connected clusters, which could represent significant modules, enriched these clusters with alternative conformations, and used PRISM to predict the interactions between the conformations. We proposed putative VCI-related interactions, such as NFKBIA-RELA and the proteasome complex, as well as their effects. The five interactions that we discovered have a higher predicted binding affinity when one of the conformations is mutated: LTF-SNCA, FGA-LTF, UBE2D1-VCP, ERBB4-INS, and NFE2L2-VCP. Additionally, since VCP has a conformational mutation linked to dementia, we proposed that the cancer-related protein BRCA1 may have implications for VCI. BRCA1s interaction with both wild-type and mutant XRCC4 and LIG4 suggests the significance of the DNA damage response pathway which can be shared between VCI and cancer.Altogether, our results suggest various pathways and interactions that can act as targets for therapeutic interventions or early diagnosis of VCI.

systems biology↗

bioRxiv2024

Comparative Analysis of Structural Alignment Algorithms for Protein-Protein Interfaces in Template-Based Docking Studies

Protein-protein interactions are pivotal for various functions within living organisms. Understanding their underlying mechanisms holds significant potential for unraveling cellular processes. There are several methods to identify protein-protein interactions, including but not limited to template-based docking. The power of template docking lies in the template library selection and the quality of structural alignment. Within the scope of our investigation, we specifically delve into the performance of four structural alignment algorithms on one protein interface and four protein structure benchmark sets. This study places particular emphasis on assessing these tools on protein interfaces, composed of non-continuous structure segments, as these interfaces play a crucial role in protein interactions, especially in the context of template-based docking. Notably, our findings indicate that TM-align, despite not being explicitly designed for sequence-order independent alignment, exhibits comparable performance to tools tailored for this purpose while executing in a considerably shorter time frame. Therefore, TM-align emerges as a promising candidate for the crucial structural alignment step in template-docking pipelines.

bioinformatics↗

bioRxiv2023

ProInterVal: Validation of Protein-Protein Interfaces through Learned Interface Representations

Proteins are vital components of the biological world, serving a multitude of functions. They interact with other molecules through their interfaces and participate in crucial cellular processes. Disruptions to these interactions can have negative effects on the organism, highlighting the importance of studying protein-protein interfaces for developing targeted therapies for diseases. Therefore, the development of a reliable method for investigating protein-protein interactions is of paramount importance. In this research, we present an approach for validating protein-protein interfaces using learned interface representations. The approach involves using a graph-based contrastive autoencoder architecture and a transformer to learn representations of proteinprotein interaction interfaces from unlabeled data, then validating them through learned representations with a graph neural network. Our method achieves an accuracy of 0.91 for the test set, outperforming existing GNN-based methods. We demonstrate the effectiveness of our approach on a benchmark dataset and show that it provides a promising solution for validating protein-protein interfaces.

bioinformatics↗

bioRxiv2023

Understanding Molecular Links of Vascular Cognitive Impairment: Selective Interaction between Mutant APP, TP53, and MAPKs

Vascular cognitive impairment (VCI) is an understudied cerebrovascular disease. As it can result in a significant amount of functional and cognitive disabilities, it is vital to reveal proteins related to it. Our study focuses on revealing proteins related to this complex disease by deciphering the crosstalk between cardiovascular and cognitive diseases. We build protein-protein interaction networks related to cardiovascular and cognitive diseases. After merging these networks, we analyze the network to extract the hub proteins and their interactors. We found the clusters on this network and built the structural protein-protein interaction network of the most connected cluster on the network. We analyzed the interactions of this network with molecular modeling via PRISM. PRISM predicted several interactions that can be novel in the context of VCI-related interactions. Two mutant forms of APP (V715M and L723P), previously not connected to VCI, were discovered to interact with other proteins. Our findings demonstrate that two mutant forms of APP interact differently with TP53 and MAPKs. Furthermore, TP53, AKT1, PARP1, and FGFR1 interact with MAPKs through their mutant conformations. We hypothesize that these interactions might be crucial for VCI. We suggest that these interactions and proteins can act as early VCI markers or as possible therapeutic targets.

systems biology↗

bioRxiv2023

DiPPI: A curated dataset for drug-like molecules in protein-protein interfaces

Proteins interact through their interfaces, and dysfunction of protein-protein interactions (PPIs) has been associated with various diseases. Therefore, investigating the properties of the drug-modulated PPIs and interface-targeting drugs is critical. Here, we present a curated large dataset for drug-like molecules in protein interfaces. We further present DiPPI (Drugs in Protein-Protein Interfaces), a two-module website to facilitate the search for such molecules and their properties by exploiting our dataset in drug repurposing studies. In the interface module of the website, we extracted several properties of interfaces, such as amino acid properties, hotspots, evolutionary conservation of drug-binding amino acids, and post-translational modifications of these residues. On the drug-like molecule side, we curated a list of drug-like small molecules and FDA-approved drugs from various databases and extracted those that bind to the interfaces. We further clustered the drugs based on their molecular fingerprints to confine the search for an alternative drug to a smaller space. Drug properties, including Lipinskis rules and various molecular descriptors, are also calculated and made available on the website to guide the selection of drug molecules. Our dataset contains 534,203 interfaces for 98,632 proteins, of which 55,135 are detected to bind to a drug-like molecule. 2,214 drug-like molecules are deposited on our website, among which 335 are FDA-approved. DiPPI provides users with an easy-to-follow scheme for drug repurposing studies through its well-curated and clustered interface and drug data; and is freely available at http://interactome.ku.edu.tr:8501.

bioinformatics↗

bioRxiv2023

Conformational diversity and protein-protein interfaces in drug repurposing in Ras signaling pathway

Ras/Raf/MEK/ERK signaling pathway regulates cell growth, division, and differentiation. In this work, we focus on drug repurposing in the Ras/Raf/MEK/ERK signaling pathway, considering structural similarities of protein-protein interfaces. The complexes in this pathway are extracted from literature and interfaces formed by physically interacting proteins are found via PRISM (PRotein Interaction by Structural Matching) if not available in Protein Data Bank. As a result, the structural coverage of these interactions has been increased from 21% to 92% using PRISM. Multiple conformations of each protein are used to include protein dynamics and diversity. Next, we find FDA-approved drugs bound to additional structurally similar protein-protein interfaces. The results suggest that HIV protease inhibitors tipranavir, indinavir and saquinavir may bind to Epidermal Growth Factor Receptor (EGFR) and Receptor Tyrosine-Protein Kinase ErbB-3 (ERBB3/HER3) interface. Tipranavir and indinavir may also bind to EGFR and Receptor Tyrosine-Protein Kinase ErbB-2 (ERBB2/HER2) interface. Additionally, a drug used in Alzheimers disease (galantamine) and an antinauseant for cancer chemotherapy patients (granisetron) can bind to RAF proto-oncogene serine/threonine-protein kinase (RAF1) and Serine/threonine-protein kinase B-Raf (BRAF) interface. Hence, we propose an algorithm to find drugs to be potentially used for cancer. As a summary, we propose a new strategy of using a dataset of structurally similar protein-protein interface clusters rather than pockets in a systematic way. Significance statementThis work focuses on drug repurposing in the Ras/Raf/MEK/ERK signaling pathway based on structural similarities of protein-protein interfaces. The Food and Drug Administration approved drugs bound to the protein-protein interfaces are proposed for the other interfaces using protein-protein interface clusters based on structural similarities. Moreover, the structural coverage of protein complexes of physical interactions in the pathway has been increased from 21% to 92% using multiple conformations of each protein to include protein dynamics.

bioinformatics↗

bioRxiv2023

Revealing Shared Proteins and Pathways in Cardiovascular and Cognitive Diseases Using Protein Interaction Network Analysis

One of the primary goals of systems medicine is detecting putative proteins and pathways involved in disease progression and pathological phenotypes. Vascular Cognitive Impairment (VCI) is a heterogeneous condition manifesting as cognitive impairment resulting from vascular factors. The precise mechanisms underlying this relationship remain unclear, which poses challenges for experimental research. Here, we applied computational approaches like systems biology to unveil and select relevant proteins and pathways related to VCI by studying the crosstalk between cardiovascular and cognitive diseases. In addition, we specifically included signals related to oxidative stress, a common etiologic factor tightly linked to aging, a major determinant of VCI. Our results show that pathways associated with oxidative stress are quite relevant, as most of the prioritized vascular-cognitive genes/proteins were enriched in these pathways. Our analysis provided a short list of proteins that could be contributing to VCI: DOLK, TSC1, ATP1A1, MAPK14, YWHAZ, CREB3, HSPB1, PRDX6, and LMNA. Moreover, our experimental results suggest a high implication of glycative stress, generating oxidative processes and post-translational protein modifications through advanced glycation end-products (AGEs). We propose that these products interact with their specific receptors (RAGE) and Notch signaling to contribute to the etiology of VCI.

systems biology↗

bioRxiv2023

Structural Coverage of the Human Interactome

Complex biological processes in cells are embedded in the interactome, representing the complete set of protein-protein interactions. Mapping and analyzing the protein structures are essential to fully comprehending these processes molecular details. Therefore, knowing the structural coverage of the interactome is essential to show the current limitations. Structural modeling of protein-protein interactions requires accurate protein structures. In this study, we mapped all experimental structures to the reference human proteome. Later, we found the enrichment in structural coverage when complementary methods such as homology modeling and deep learning (AlphaFold) are included. We then collected the interactions from the literature and databases to form the reference human interactome resulting in 117,897 non-redundant interactions. When we analyzed the structural coverage of the interactome, we found that the number of experimentally determined protein complex structures is scarce, corresponding to 3.95% of all binary interactions. We also analyzed known and modeled structures to potentially construct the structural interactome with a docking method. Our analysis showed that 12.97% of the interactions from HuRI, 73.62%, and 32.94% from the filtered versions of STRING and HIPPIE could potentially be modeled with a high structural coverage or accuracy, respectively. Overall, this paper provides an overview of the current state of structural coverage of the human proteome and interactome. Significance StatementWe gathered binary protein-protein interactions from three prominent interactome databases to create a comprehensive human reference interactome. We quantified the structural coverage of the human interactome using already available structural data from four different sources. We further evaluate the percentage of interactions that can be accurately predicted using docking methods.

bioinformatics↗

bioRxiv2020

The structural basis of the oncogenic mutant K-Ras4B homodimers

Ras proteins activate their effectors through physical interactions in response to the various extracellular stimuli at the plasma membrane. Oncogenic Ras forms dimer and nanoclusters at the plasma membrane, boosting the downstream MAPK signal. It was reported that K-Ras4B can dimerize through two major interfaces: (i) the effector lobe interface, mapped to Switch I and effector binding regions; (ii) the allosteric lobe interface involving 3 and 4 helices. Recent experiments showed that constitutively active, oncogenic mutant K-Ras4BG12D dimers are enriched in the plasma membrane. Here, we perform molecular dynamics simulations of K-Ras4BG12D homodimers aiming to quantify the two major interfaces in atomic level. To examine the effect of mutations on dimerization, two double mutations, K101D/R102E on the allosteric lobe and R41E/K42D on the effector lobe interfaces were added to the K-Ras4BG12D dimer simulations. We observed that the effector lobe K-Ras4BG12D dimer is stable, while the allosteric lobe dimer alters its helical interface during the simulations, presenting multiple conformations. The K101D/R102E mutations slightly weakens the allosteric lobe interface. However, the R41E/K42D mutations disrupt the effector lobe interface. Using the homo-oligomers prediction server, we obtained trimeric, tetrameric, and pentameric complexes with the allosteric lobe K-Ras4BG12D dimers. However, the allosteric lobe dimer with the K101D/R102E mutations is not capable of generating multiple higher order structures. Our detailed interface analysis may help to develop inhibitor design targeting functional Ras dimerization and high order oligomerization at the membrane signaling platform.

biophysics↗

bioRxiv2019

DeepInterface: Protein-protein interface validation using 3D Convolutional Neural Networks

MotivationProtein-protein interactions are crucial in almost all biological processes. Proteins interact through their interfaces. It is important to determine how proteins interact through interfaces to understand protein binding mechanisms and to predict new protein-protein interactions.\n\nResultsWe present DeepInterface, a deep learning based method which predicts, for a given protein complex, if the interface between the proteins of a complex is a true interface or not. The model is a 3-dimensional convolutional neural networks model and the positive datasets are obtained from all complexes in the Protein Data Bank, the negative datasets are the incorrect solutions of the docking decoys. The model analyzes a given interface structure and outputs the probability of the given structure being an interface. The accuracy of the model for several interface data sets, including PIFACE, PPI4DOCK, DOCKGROUND is approximately 88% in the validation dataset and 75% in the test dataset. The method can be used to improve the accuracy of template based PPI predictions.

bioinformatics↗

bioRxiv2019

3D spatial organization and network-guided comparison of mutation profiles in Glioblastoma reveals similarities across patients

Mutation profiles of Glioblastoma (GBM) tumors are very heterogeneous which is the main challenge in the interpretation of the effects of mutations in disease. Additionally, the impact of the mutations is not uniform across the proteins and protein-protein interactions. The pathway level representation of the mutations is very limited. In this work, we approach these challenges through a systems level perspective in which we analyze how the mutations in GBM tumors are distributed in protein structures/interfaces and how they are organized at the network level. Our results show that out of 14644 mutations, 4392 have structural information and ~13% of them form spatial patches. Despite a small portion of all mutations, 3D patches partially decrease the heterogeneity across the patients. Hub proteins adapt multiple patches of mutations usually with a very large one and connects mutations in multiple binding sites through the core of the protein. We reconstructed patient specific networks for 290 GBM tumors. Network-guided analysis of mutations completes the interaction components that mutated proteins potentially affect, and groups the patients according to the reconstructed networks. As a result, we found 4 tumor clusters that overcome the heterogeneity in mutation profiles, and reveal predominant pathways in each group. Additionally, the network-based similarity analysis shows that each group of patients carries a set of signature 3D mutation patches. We believe that this study provides another perspective to the analysis of mutation effects and a good training towards the network-guided precision medicine.\n\nAuthor SummaryGlioblastoma (GBM) is the most aggressive brain tumor type with a 15 months of survival on average. The mutation distribution of the GBM patients is very heterogeneous and standard treatments fail to consider the inter-tumor heterogeneity. In our study, we follow a systems level approach that integrates mutation profiles with protein-protein interaction networks. We hypothesized that although the mutations are heterogeneous, the mutations that are close in 3D of the same protein may affect the protein function similarly and this information can be used to get meaningful relation between disease state and mutations. Therefore, we spatially grouped these mutations as \"patches\" and reconstructed patient specific protein interaction networks. When we cluster these networks based on their pathway similarities, we found four patient groups in GBM. Then, the comparison of groups revealed overrepresentation of 3D patches. This finding can be used for patient specific therapy.

systems biology↗