Dr. Shuvo was recently awarded a PVAMU RISE Undergraduate Award, “Toward Trustworthy and End-to-End AI for Next-Generation Computational Therapeutics”, to develop a trustworthy, explainable, and secure end-to-end AI framework that integrates multiple computational therapeutics tasks—from protein function prediction to drug response prediction—to accelerate and improve drug discovery. This is a 1-year grant to support an undergraduate student, up to $7,200.

Project Title: Toward Trustworthy and End-to-End AI for Next-Generation Computational Therapeutics

Principal Investigator: Md Hossain Shuvo, Ph.D.

Project Description:

Drug discovery is a long and expensive process, with many potential drug candidates failing before reaching clinical use. Artificial intelligence can help reduce this burden by identifying and prioritizing promising drug candidates earlier, but most existing AI tools focus on individual tasks and do not work together as a connected system. This project will develop a modular, end-to-end AI framework that connects important computational therapeutics tasks, including protein function prediction, drug–target affinity prediction, toxicity prediction, and drug response prediction. The project will also incorporate self-assessment and explainability so that the system can provide confidence estimates, identify important features behind its predictions, and better understand when and why models fail. In addition, the project will examine the effects of low-quality and potentially adversarial inputs and develop a secure deployment prototype with basic safeguards such as input validation, access control, secure communication, and activity logging. The expected outcome is a working prototype that demonstrates how multiple AI models can be connected into a more reliable, transparent, and secure computational therapeutics workflow. The results will provide a foundation for future research on trustworthy AI for biomedical applications and support the development of larger NIH, NSF, and DoD proposals.

Sponsor: R&I Division at Prairie View A&M University Funding Amount: $7,200 Project Period: FY 2026–2027

Dr. Shuvo was recently awarded an NSF Research Initiation Award: Bridging the Gap between Sequence and Function with Ethical and Explainable AI for Protein Function Prediction. This is a 3-year, $445,686 grant, sponsored by DUE (Division Of Undergraduate Education).

Project Title: Bridging the Gap between Sequence and Function with Ethical and Explainable AI for Protein Function Prediction

Principal Investigator: Md Hossain Shuvo, Ph.D.

Project Description:

This project will develop advanced deep learning frameworks for protein function prediction by integrating protein sequence, three-dimensional structural, protein–protein interaction, and protein–nucleic acid interaction information within ethical and explainable AI models.

The research will explore transformer models, graph neural networks, convolutional neural networks, and hybrid learning approaches to improve the prediction of biological functions from protein sequences. The project will emphasize explainability, transparency, and robustness and will evaluate the developed methods using publicly available benchmark datasets and the Critical Assessment of Functional Annotation (CAFA) Challenge.

The project will also provide open-source software, web-based computational tools, and educational materials to support the scientific community, while expanding student research and workforce development opportunities.

NSF Award Details

Dr. Victoria Mgbemena recently led a team of multidisciplinary researchers (Dr. Kim - PVAMU, Dr. Ali - TAMU, Dr. Darwish - AgriLife) to earn a TAMUS REF Early Stage Development Grant. This is a 1-year, $100,000, TAMUS internal grant to significantly elevate research capacity, competitiveness, and impact across the A&M System.

Project Title: Targeting and Characterization of Alzheimer’s Disease by a Novel Fyn Kinase Inhibitor

Principal Investigator: Victoria Mgbemena, Ph.D. Co-PIs:

• Dr. Seungchan Kim (Co-PI, PVAMU)
• Dr. Hamed I. Aly Ismail (Co-PI, TAMU)
• Dr. Ahmed Darwish (Co-PI, AgriLife)

Project Summary:

Overview of Proposed Research Concept: Alzheimer’s disease (AD) is a progressive neurodegenerative disorder and the leading cause of dementia, accounting for 60-80% of cases and projected to affect ∼13 million Americans by 2050 (Kelliny S. et al., 2025, Alzheimer’s Association, 2023,Centers for Disease Control and Prevention [CDC], 2024, Alzheimer’s Association, 2025). Despite recent advances in amyloid-targeting antibodies, including lecanemab and donanemab (Tassinari & Milelli, 2025), no therapy effectively reverses the synaptic dysfunction underlying cognitive decline (Daly et al., 2024, Papaliagkas, 2025, Kim et al., 2025, Wang et al., 2025). Thus, innovative mechanisms that directly target early synaptic failure are critically needed. AD neuropathology is characterized by amyloid-β (Aβ) plaques, tau neurofibrillary tangles (TNFs), synaptic loss, and neuroinflammation (Bloom, 2014). A key pathway driving synaptic dysfunction in AD is the tau/Fyn/N-methyl-D-aspartate receptor (NMDAR) axis. Under normal conditions, tau recruits the Src-family kinase Fyn to postsynaptic sites to modulate NMDAR signaling (Fan et al., 2022). The Fyn kinase belongs to a family of proteins important for signal transduction across multiple cellular pathways, including immune cell activation of T-cells and B-cells (Thomas & Brugge, 1997; Salmond et al., 2009; Banerjee et al., 2013). In AD, mislocalized tau aberrantly scaffolds Fyn within dendritic spines (Ittner et al., 2010; Chin et al., 2005), promoting GluN2B-Y1472 phosphorylation and driving NMDAR over-activation. This triggers $\text{Ca}^{2+}$-dependent excitotoxicity, synaptic dysfunction, tau hyperphosphorylation, and Aβ secretion (Fan et al., 2022, Peng & Fu, 2023; van Dyck et al., 2019; Guglietti et al., 2021). Genetic and in vivo evidence show that tau-dependent Fyn signaling functionally links Aβ-toxicity to synaptic collapse and AD pathology. While amyloid burden shows weak correlation with clinical symptoms, tau pathology and synaptic degeneration closely track with disease severity (Bloom, 2014; Fan et al., 2022; Thomas & Brugge, 1997, Ittner et al., 2010), highlighting synapses as high-value therapeutic targets. This project targets Fyn kinase, a non-receptor tyrosine kinase and key mediator of synaptic toxicity induced by Aβ and Tau. There is currently a noteworthy potential drug which targets Fyn kinase known as AZD0530, saracatinib. Saracatinib is currently being repurposed for efficacy testing in AD in preclinical and clinical tests. So far in these tests, it has been shown to rescue memory, restore synaptic density and reduce microglial activation in AD mouse models (Martínez-Mármol et al., 2023; Yadikar et al., 2020). Dr. Hamed Ali’s team has identified a high-potential small-molecule candidate (AB157) with high predicted blood-brain barrier (BBB) penetrability. We will test the lead compound using a human 3D vascularized forebrain organoid model. By integrating iPSC-derived neural/mesenchymal aggregates, we will create a physiologically relevant “humanized” system to evaluate the reduction of pathogenic Aβ and Tau phosphorylation without systemic cytotoxicity. The goal of this project is to investigate and understand molecular shifts over time in the presence and absence of our team’s Fyn kinase inhibitor.

Alignment with State, National Research Priorities: Our research aligns directly with the National Institute on Aging (NIA) and National Institutes of Health (NIH) goals to increase the development of effective treatments for AD and related dementias (ADRD). Our project addresses a local public health challenge relating to an aging population, and targets specific biological mechanisms of Aging and Dementia. It also aligns well with strategic goals of the Dementia Prevention Research Institute of Texas (DPRIT).

Anticipated Impact: It is crucial for society to investigate and understand the etiology of AD , so that prevention and treatment strategies can lessen the burden on families and the health care system. The successful completion of this project will provide an effective pipeline construct for drug screening, testing and evaluation, and provide a pathway between in vitro modeling and preclinical mouse studies.

Abstract

The rapid evolution of high-throughput sequencing has transformed our ability to profile biological systems, yet the inherent relational complexity of these data necessitates advanced computational frameworks. This seminar explores the paradigm shift from traditional deep learning to Graph Neural Networks (GNNs), a class of models uniquely equipped to handle non-Euclidean data. We begin by establishing the mathematical foundations of GNNs, focusing on the message-passing framework and the role of graph convolutional operators in aggregating neighborhood information. We will contrast standard architectures—such as Graph Convolutional Networks (GCNs) and Graph Attention Networks (GATs)—while highlighting their utility in both general domains (social networks, recommendation engines) and specialized biomedical research.

The core of the presentation delves into the integration of GNNs with single-cell RNA sequencing (scRNA-seq) and spatial transcriptomics. We will discuss how cell-cell similarity graphs and physical spatial coordinates can be modeled to perform critical tasks such as cell-type annotation, data imputation, and spatial domain detection. Furthermore, we will examine the application of GNNs in reconstructing Gene Regulatory Networks (GRNs). By treating genes as nodes and regulatory interactions as edges, GNNs provide a robust mechanism for inferring causal relationships and identifying spatially varying regulatory patterns that are often obscured in traditional, non-graph-based analyses.

The seminar will also review the current ecosystem of tools, including PyTorch Geometric (PyG) and specialized packages like SpaGCN and DeepSEM. We conclude by discussing future directions for GNNs in modeling multi-modal biomedical data.

Speaker Bio

Dr. Seungchan Kim is a Chief Scientist and Executive Professor at the Department of Electrical and Computer Engineering and the Director of the CRI Center for Computational Systems Biology at the Prairie View A&M University (PVAMU). Prior to this appointment, he was the Head of Biocomputing Unit and an Associate Professor at Integrated Cancer Genomics Division of Translational Genomics Research Institute (TGen). He was one of the founding faculty members of TGen, founded in 2002, by Dr. Trent, then-Scientific Director of the National Human Genome Research Institute at the National Institutes of Health, leading computational systems biology research at the institute. He was also an Assistant Professor in the School of Computing, Informatics, Decision Systems Engineering (CIDSE) at the Arizona State University from 2004 till 2011. Dr. Kim received B.S. and M.S. degrees in Agriculture Engineering from the Seoul National University, and Ph.D. in Electrical Engineering from the Texas A&M University. He also got his post-doctoral training at the Cancer Genetics Branch of National Human Genome Research Institute.

Dr. Kim’s research interests include: 1) mathematical modeling of genetic regulatory networks, 2) development of computational methods to analyze multitude of high throughput multi-omics data to identify disease biomarkers, and 3) computational models to diagnose patients or predict patient outcomes, for example, disease subtypes or drug response. His studies have had a large influence on the development of computational tools to study underlying mechanisms for cancer development and better understand the molecular mechanisms behind cancer biology and biological systems.

Abstract

Cybercrime is often approached as a technical problem, but its global distribution reflects broader social, institutional, and governance conditions. This talk examines cybercrime as a socio-technical system through two complementary studies. First, using FireHOL IP blocklist data, I compare a Generalized Linear Model (GLM) with several non-linear machine learning approaches to identify the global drivers of cybercrime. The Random Forest Regressor achieves the strongest predictive performance, showing that cybercrime is shaped by complex, non-linear relationships that are not well captured by conventional linear approaches. Socio-economic and governance variables—including poverty rate, adult population, government effectiveness, and rule of law—emerge as especially important predictors, highlighting that cybercrime is not driven by technical infrastructure alone. Second, I examine how U.S. criminology and criminal justice doctoral programs are preparing scholars to study this increasingly complex domain. A review of 43 Ph.D. programs shows that only about half offer cybercrime-related coursework, with just six maintaining dedicated laboratories. Cybercrime is typically treated as an elective and often lacks technical or interdisciplinary depth. Taken together, these findings reveal a mismatch between the complexity of cybercrime and the current structure of doctoral training. The talk argues for a more integrated approach that connects computational methods, social science theory, and interdisciplinary training to better understand and respond to cybercrime.

Speaker Bio

Dr. Ling Wu is an Associate Professor in the Department of Criminology and Criminal Justice at the University of Alabama. Her research focuses on cybercrime, cyber victimization, and the social and structural dimensions of digital crime. Her work applies quantitative and computational approaches to examine how technological, social, and institutional factors shape cybercrime risks and patterns.