Biology subjects

Qazi, S.

Publications and source records attributed to Qazi, S..

6 recordsLinked to original sources

bioRxiv2026

MetaResNet: Enhancing Microbiome-Based Disease Classification through Colormap Optimization and Imbalance Handling

Image-based representations of metagenomic data enable convolutional neural network (CNN) applications for microbiome disease classification. However, the impact of colormap selection on model performance remains unexplored. Current approaches arbitrarily select visualization parameters despite evidence that colormap choices can suppress minority-class features in imbalanced microbiome datasets. This study systematically evaluates colormap effects on metagenomic disease classification to establish evidence-based visualization guidelines. We developed MetaResNet, a custom CNN architecture incorporating residual blocks and attention mechanisms, to assess five colormap schemes (Jet, YlGnBu, Reds, Paired and nipy spectral) across four benchmark datasets: inflammatory bowel disease (n=110), colon cancer (n=121), women type 2 diabetes (n=96), and obesity (n=253). Class imbalance was addressed using Synthetic Minority Over-sampling Technique (SMOTE) versus class weighting strategies. Custom data augmentation preserved taxonomic abundance relationships while enhancing generalization. Performance evaluation employed F1-score, Receiver Operating Characteristic and Area Under the Curve (AUC-ROC), Matthews correlation coefficient (MCC), precision, and recall to address accuracy limitations in imbalanced scenarios. Results identified the Jet colormap coupled with SMOTE as the optimal global configuration, maximizing signal retention and achieving peak performance (AUC 1.00 in Colon). SMOTE significantly improved minority-class recall over class weighting (0.81 {+/-} 0.09 vs. 0.69 {+/-} 0.11, p = 0.003). MetaResNet achieved performance comparable to current state-of-the-art frameworks, while statistically outperforming established deep learning baselines (e.g., DeepMicro, PopPhy-CNN; p = 0.025) in discriminatory power (AUC), with peak values exceeding 0.96. These findings demonstrate that visualization efficacy is strategy-dependent, establishing MetaResNet as a robust framework for microbiome-based diagnostics that supports evidence-based visualization strategies for precision medicine.

bioinformatics↗

bioRxiv2026

DynaBiomeX: An Interpretable Dual-Strategy Deep Learning Framework for Architectural Noise Filtration in Sparse Longitudinal Microbiome Data

ObjectiveLongitudinal microbiome datasets present unique challenges due to extreme sparsity, zero-inflation, and non-stationary behavior. Conventional Recurrent Neural Networks (RNNs) struggle to distinguish structural from sampling zeros in these contexts, limiting their utility for Clinical Decision Support (CDS). MethodsWe introduce DynaBiomeX, an interpretable framework specifically developed for sparse biomedical time-series. It integrates Stacking Ensembles (Bi-LSTM, GRU) with an adapted Temporal Fusion Transformer (TFT) in a unified Screener-Sentinel workflow. The Ensembles optimize collective decision boundaries to maximize sensitivity and minimize missed cases. Concurrently, the TFT functions as a Physiological Gatekeeper, utilizing Gated Residual Networks (GRN) to actively filter stochastic noise from real biological signals. We validated this approach on a multi-modal dataset of 1,871 hematopoietic cell transplantation (HCT) patients to detect gut dysbiosis. ResultsStacking ensembles maximized discriminative performance (ROC-AUC = 0.912), effectively serving as high-sensitivity screeners. In contrast, the Adapted TFT functioned as a precision Sentinel, achieving zero false positives (Precision = 1.0) and high stability (MCC = 0.646). Crucially, the TFT demonstrated superior probabilistic reliability with a low Expected Calibration Error (ECE = 0.0085), addressing the "black-box" overconfidence typical of deep learning models. Ablation studies confirmed predictive robustness even without clinical covariates (ROC-AUC > 0.81). ConclusionDynaBiomeX couples sensitive screening with precise, calibrated validation to robustly analyze sparse longitudinal data. Validated on microbiome dysbiosis, this framework offers a scalable template for zero-inflated domains like single-cell sequencing and EHR monitoring. HighlightsO_LIDynaBiomeX integrates Stacking Ensembles and Temporal Fusion Transformers for microbiome risk stratification. C_LIO_LIGated Residual Networks distinguish stochastic sampling zeros from structural zeros in sparse ASV data. C_LIO_LIA novel Screener-Sentinel workflow resolves the trade-off between surveillance sensitivity and diagnostic precision. C_LIO_LIClinical metadata functions as a physiological gatekeeper, filtering noise to achieve zero false positives (MCC 0.646). C_LIO_LIThe framework minimizes alarm fatigue in Auto-FMT decision support by validating latent dysbiosis signals. C_LI

bioinformatics↗

bioRxiv2024

IDENTIFICATION OF KEY GENES ASSOCIATED WITH PCOS (POLYCYSTIC OVARIAN SYNDROME) AND ENDOMETRIAL/OVARIAN CANCER THROUGH BIOINFORMATICS

BackgroundPCOS, a common endocrine disorder, is linked to increased risks of endometrial cancer (EC) and ovarian cancer (OC). Our study utilizes bioinformatics analysis to identify shared gene signatures and elucidate biological processes between EC/OC and PCOS. Materials and MethodsGene expression profiles for PCOS (GSE199225), endometrial cancer (GSE215413), and ovarian cancer (GSE174670) were obtained from the GEO database. Hub genes were identified through functional enrichment analysis and protein-protein interaction. Drug identification analyses were employed to find drugs targeting the hub genes. ResultKey hub genes linking PCOS and EC includes RECQL4, RAD54L, ATR, CHTF18, WDHD1, CDT1, PLK1, PKMYT1, RAD18, and RPL3; for PCOS and OC, they include HMOX1, TXNRD1, NQO1, GCLC, GSTP1, PRDX1, SOD1, GPX3, BOP1, and BYSL. Gene Ontology analysis revealed DNA Metabolic Processes in PCOS and EC, while in PCOS and OC, it identified the Removal of Superoxide Radicals. KEGG Pathway analysis highlighted Cell cycle in PCOS and EC and Hepatocellular carcinoma in PCOS and OC. Potential drugs for PCOS and EC include quercetin, calcitriol, and testosterone; for PCOS and OC, eugenol and 1-chloro-2,4-dinitrobenzene are identified. ConclusionThese findings offer insights into potential therapeutic targets and pathways linking PCOS with EC and OC, enhancing our understanding of the molecular mechanisms involved in these associations.

cancer biology↗

bioRxiv2024

Impact of Capillary and Sarcolemmal Proximity on Mitochondrial Structure and Energetic Function in Skeletal Muscle

Mitochondria within skeletal muscle cells are considered to be located either between the muscle contractile apparatus (interfibrillar mitochondria, IFM) or in large pools beneath the cell membrane (subsarcolemmal mitochondria, SSM), with several structural and functional differences reported between IFM and SSM. However, recent 3D imaging studies suggest that proximity to capillaries embedded in sarcolemmal grooves, rather than proximity to the sarcolemma itself, may drive the accumulation of mitochondria near the cell periphery (paravascular mitochondria, PVM). To evaluate the impact of capillary versus sarcolemmal proximity, we compared the structure and function of skeletal muscle mitochondria located either in large pools lateral to embedded capillaries (PVM), adjacent to the sarcolemma but not in PVM pools (SSM), or interspersed between sarcomeres (IFM). Mitochondrial morphology and interactions were assessed by 3D electron microscopy coupled with machine learning segmentation while mitochondrial energy conversion was assessed by two-photon microscopy of mitochondrial membrane potential, content, calcium, NADH redox and flux in live, intact cells. Structurally, while PVM and SSM were similarly larger than IFM, PVM were more compact and had greater mitochondrial connectivity compared to both IFM and SSM. Functionally, PVM had similar or greater basal NADH flux compared to SSM and IFM, respectively, despite a more oxidized NADH pool and a greater membrane potential, signifying a greater activation of the electron transport chain in PVM. Together, these data indicate proximity to capillaries has a greater impact on mitochondrial energy conversion and distribution in skeletal muscle than the sarcolemma alone.

physiology↗

bioRxiv2021

BESFA: Bioinformatics based Evolutionary, Structural & Functional Analysis of Prostrate, Placenta, Ovary, Testis, and Embryo (POTE) Paralogs

The POTE family comprises 14 paralogues and is primarily expressed in Prostrate, Placenta, Ovary, Testis, Embryo (POTE), and cancerous cells. The prospective function of the POTE protein family under physiological conditions is less understood. We systematically analyzed their cellular localization and molecular docking analysis to elucidate POTE proteins structure, function, and Adaptive Divergence. Our result discerns that group three POTE paralogs (POTEE, POTEF, POTEI, POTEJ, and POTEKP (a pseudogene)) exhibits significant variation among other members could be because of their Adaptive Divergence. Furthermore, our molecular docking studies on POTE protein revealed the highest binding affinity with NCI-approved anticancer compounds. Additionally, POTEE, POTEF, POTEI, and POTEJ were subject to an explicit molecular dynamic simulation for 50ns. MM-GBSA and other essential electrostatics were calculated that showcased that only POTEE and POTEF have absolute binding affinities with minimum energy exploitation. Thus, this studys outcomes are expected to drive cancer research to successful utilization of POTE genes family as a new biomarker, which could pave the way for the discovery of new therapies.

cancer biology↗

bioRxiv2020

Virion-associated spermidine transmits with Rift Valley fever virus particles to maintain infectivity

Viruses require host cell metabolites to productively infect, and the mechanisms by which viruses usurp these molecules is diverse. One group of cellular metabolites important in virus infection is the polyamines, small positively-charged molecules involved in cell cycle, translation, and nucleic acid synthesis, among other cellular functions. Polyamines also support replication of diverse viruses, and they are important for processes such as transcription, translation, and viral protein enzymatic activity. Rift Valley fever virus (RVFV) is a negative-sense RNA virus that requires polyamines to produce infectious particles. In polyamine depleted conditions, noninfectious particles are produced that interfere with virus replication and stimulate immune signaling. Here, we find that RVFV relies on virion-associated polyamines to maintain infectivity. We show that RVFV replication is facilitated by any of the three biogenic polyamines; however, we specifically find spermidine associated with purified virions. Using a panel of polyamine homologs, we observe that virions can also associate with (R)-3-methylspermidine and norspermidine, though not with other less homologous molecules. Using polyamine reporter cells, we demonstrate that virion-associated polyamines transmit from one infected cell to another. Finally, we find that virions devoid of polyamines are unstable and cannot be supplemented with exogenous polyamines to regain stability or infectivity. These data highlight a unique role for polyamines, and spermidine in particular, in maintaining virus infectivity, a function not previously appreciated. Further, these studies are the first to identify polyamines associated with RVFV virions. Targeting polyamines represents a promising antiviral strategy, and this work highlights a new mechanism by which we can inhibit virus replication through FDA-approved polyamine depleting pharmaceuticals.

microbiology↗