Biology subjects

Akhter, S.

Publications and source records attributed to Akhter, S..

5 recordsLinked to original sources

bioRxiv2025

An Innovative Framework for Heart Sound Classification Integrating Adaptive Fuzzy Rank-Based Ensemble of Transfer Learning Models with Multi-Dimensional Features

Heart sound identification for diagnosing cardiovascular diseases is a complex challenge due to the intricate, spectro-temporal characteristics of conditions like Aortic Stenosis, Mitral Regurgitation, and multi-valvular disorders. Conventional methods frequently inadequately encompass the complete range of diagnostic characteristics, depending on fragmented or simplistic approaches that lack generalizability across varied patient demographics, clinical settings, and recording circumstances. To address these limitations, we offer an innovative, multi-dimensional feature fusion framework that integrates Wavelet Scattering Transform (WST) and Mel-Frequency Cepstral Coefficients (MFCC) to capture both temporal stability and perceptually optimal frequency patterns from cardiac sounds. This method is further refined by an adaptive fuzzy rank-based ensemble technique utilizing the Gompertz function, which dynamically modifies model weights according to confidence metrics, hence assuring more dependable and precise predictions amidst fluctuating clinical uncertainty. We meticulously assess our model utilizing eight advanced fine-tuned transfer learning architectures across four feature extraction techniques (WST, MFCC, STFT, Multi-Dimensional) on the clinically validated BUET Multi-disease Heart Sound (BMD-HS) dataset. This dataset comprises 864 phonocardiogram recordings from 108 participants with echocardiographically confirmed diagnoses. The multi-dimensional feature fusion method attains 97% accuracy with EfficientNetB2, whereas the adaptive fuzzy ensemble strategy achieves 98% accuracy, surpassing both individual models and conventional ensemble methods. Moreover, Explainable AI with Audio-LIME offers transparent, clinically interpretable insights, achieving fidelity scores surpassing 0.85 and clinical relevance ratings exceeding 90%, facilitating the identification of critical time-frequency regions that hold diagnostic significance. This system establishes a novel benchmark for heart sound classification, enhances the differentiation of intricate valvular disorders, and provides dependable, evidence-based decision support for cardiovascular diagnosis in clinical settings. Our research illustrates the capability for resilient, scalable, and interpretable heart sound classification systems that can improve clinical decision-making and promote the integration of automated diagnostic tools in healthcare.

bioengineering↗

bioRxiv2025

Bacteriocin Prediction Through Cross-Validation-Based and Hypergraph-Based Feature Evaluation Approaches

Bacteriocins offer a promising solution to antibiotic resistance, possessing the ability to target a wide range of bacteria with precision. Thus, there is an urgent need for a computational model to predict new bacteriocins and aid in drug development. This work centers on constructing predictive models with XGBoost machine learning algorithm, using physicochemical structural properties and sequence profiles of protein sequences. We employed correlation analyses, cross-validation, and hypergraph-based techniques to select features. Cross-validation feature evaluation (CVFE) partitions the dataset, selects features within each partition, and identifies common features, ensuring representativeness. On the contrary, hypergraph-based feature evaluation (HFE) focuses on minimizing hypergraph cut conductance, leveraging higher-order data relationships to precisely utilize information regarding feature and sample correlations. The XGBoost models were built using the selected features obtained from these two feature evaluation methods. Our HFE-based approach achieved 99.11% accuracy and an AUC of 0.9974 on the test data, overall outperforming the CVFE-based feature evaluation method and yielding results comparable to existing approaches. We also analyzed the feature contributions directly from the best model using SHapley Additive exPlanations (SHAP). Our web application, accessible at https://shiny.tricities.wsu.edu/bacteriocin-prediction/, offers prediction results, probability scores, and SHAP plots using both cross-validation- and hypergraph-based methods, along with previously implemented approaches for feature selection.

bioinformatics↗

bioRxiv2024

Mechanistic Studies of Small Molecule Ligands Selective to RNA Single G Bulges

Small-molecule RNA binders have emerged as an important pharmacological modality. A profound understanding of the ligand selectivity, binding mode, and influential factors governing ligand engagement with RNA targets is the foundation for rational ligand design. Here, we report a novel class of coumarin derivatives exhibiting selective binding affinity towards single G RNA bulges. Harnessing the computational power of all-atom Gaussian accelerated Molecular Dynamics (GaMD) simulations, we unveiled a rare minor groove binding mode of the ligand with a key interaction between the coumarin moiety and the G bulge. This predicted binding mode is consistent with results obtained from structure-activity-relationship (SAR) studies and transverse relaxation measurements by NMR spectroscopy. We further generated 444 molecular descriptors from 69 coumarin derivatives and identified key contributors to the binding events, such as charge state and planarity, by lasso (least absolute shrinkage and selection operator) regression. Strikingly, small structure perturbations on these key contributors, such as the addition of a methyl group that disrupts the planarity of the ligand resulted in > 100-fold reduction in the binding affinity. Our work deepened the understanding of RNA-small molecule interactions and integrated a new generalizable platform for the rational design of selective small-molecule RNA binders.

biochemistry↗

bioRxiv2024

Characterization of natural product inhibitors of quorum sensing in Pseudomonas aeruginosa reveals competitive inhibition of RhlR by ortho-vanillin

Quorum sensing (QS) is a cell-cell signaling system that enables bacteria to coordinate population density-dependent changes in behavior. This chemical communication pathway is mediated by diffusible N-acyl L-homoserine lactone signals and cytoplasmic signal-responsive LuxR-type receptors in Gram-negative bacteria. As many common pathogenic bacteria use QS to regulate virulence, there is significant interest in disrupting QS as a potential therapeutic strategy. Prior studies have implicated the natural products salicylic acid, cinnamaldehyde and other related benzaldehyde derivatives as inhibitors of QS in the opportunistic pathogen Pseudomonas aeruginosa, yet we lack an understanding of the mechanisms by which these compounds function. Herein, we evaluate the activity of a set of benzaldehyde derivatives using heterologous reporters of the P. aeruginosa LasR and RhlR QS signal receptors. We find that most tested benzaldehyde derivatives can antagonize LasR or RhlR reporter activation at micromolar concentrations, although certain molecules also caused mild growth defects and nonspecific reporter antagonism. Notably, several compounds showed promising RhlR or LasR specific inhibitory activities over a range of concentrations below that causing toxicity. ortho-Vanillin, a previously untested compound, was the most promising within this set. Competition experiments against the native ligands for LasR and RhlR revealed that ortho-vanillin can interact competitively with RhlR but not with LasR. Overall, these studies expand our understanding of benzaldehyde activities in the LasR and RhlR receptors and reveal potentially promising effects of ortho-vanillin as a small molecule QS modulator against RhlR. IMPORTANCE Quorum sensing (QS) regulates many aspects of pathogenesis in bacteria and has attracted interest as a target for anti-virulence therapies. As QS is regulated by low molecular weight chemical signals, the development of chemical strategies that can interfere with this cell-cell communication pathway has seen considerable scrutiny over the past 25 years. Much of this research has focused on common human pathogens, including the LasR and RhlR QS receptors in Pseudomonas aeruginosa. Potent and selective chemical agents capable of blocking the activity of these receptors remain relatively scarce, however. Natural products have provided a bounty of chemical scaffolds with anti-QS activities, but their molecular mechanisms are poorly characterized. The current study serves to fill this void by examining the activity of an important and wide-spread class of natural product QS modulators, benzaldehydes and related derivatives, in LasR and RhlR. We demonstrate that ortho-vanillin can act as a competitive inhibitor of RhlR, a receptor that has emerged and may supplant LasR in certain settings as a target for QS control in P. aeruginosa. The results and insights provided herein will advance the design of chemical tools to study QS with improved activities and selectivities.

microbiology↗

bioRxiv2022

Optimal feature selection and software tool development for bacteriocin prediction

Antibiotic resistance is a major public health concern around the globe. As a result, researchers always look for new compounds to develop new antibiotic drugs for combating antibiotic-resistant bacteria. Bacteriocin becomes a promising antimicrobial agent to fight against antibiotic resistance, due to its narrow killing spectrum. Sequence matching methods are widely used to identify bacteriocins by comparing them with the known bacteriocin sequences; however, these methods often fail to detect new bacteriocin sequences due to sequences high diversity. The ability to use a machine learning approach can help find new highly dissimilar bacteriocins for developing highly effective antibiotic drugs. The aim of this work is to identify optimal sets of features and develop a machine learning-based software tool for predicting bacteriocin protein sequences with high accuracy. We extracted potential features from known bacteriocin and non-bacteriocin sequences by considering the physicochemical and structural properties of the protein sequences. Then we reduced the feature set using statistical justifications and recursive feature elimination technique. Finally, we built support vector machine (SVM) and random forest (RF) models using the selected features and our models can achieve accuracy up to 95.54%. We compared the performance of our method with a popular sequence matching-based approach and a deep learning-based method. We also developed a software tool called Bacteriocin Prediction (BacPred) that implements the prediction model using the optimal set of features obtained from this study. The software package and its user manual are available at https://github.com/suraiya14/ML_bacteriocins/BacPred.

bioinformatics↗