Supervisors: Dr. Tanvir Hossain, Assistant Professor; Papia Rahman, Lecturer

Responsibilities and Experience:

• Applied Neural Network Framework (VAE) to train the big integrated data of 26 transcriptomic datasets for Oropharyngeal Carcinoma.
• Implemented Integrated Gradients, ensemble latent feature learning, and TCGA survival benchmarking.
• Synthesized Ag‑deposited Ni/rGO nanozymes; evaluating their peroxidase mimicking activity for monitoring neuronal oxidative stress.

Projects:

• Beyond Differential Expression: Deep Neural Profiling reveals RAP1GAP2 as a latent regulator of tumor invasion in Oropharyngeal Carcinoma.
• Synthesis and application of Ag-deposited Ni/rGO nanozymes for Oxidative Stress and Neuroinflammation Sensing.

GitHub:

• Deep-Neural-Profiling

Advisor: Dr. Tanvir Hossain, Assistant Professor

Responsibilities and Experience:

• Conducted bulk ncRNA-seq analysis on cancer cell lines, including ADMSC, BMMSC, HeLa, MCF7, MDAMB231, TM6, A549, H1975.
• Analyzed the expression profiles of Y and U miRNAs to examine their significance in extracellular vesicles (EVs), epithelial–mesenchymal transition (EMT) and in lung cancer.
• Filtered ncRNAs using ncRNAtools (RNAcentral API) and applied RFE-RF for feature selection in ML analysis.
• Discovered snc-markers of EMT, EV, and lung cancer and common among all and analyzed their qRT-PCR validation result.

Project:

• Chip development for small non-coding RNA (sncRNA) isolation and marker-based cancer detection.

GitHub:

• small-non-coding-RNA

Supervisor: Dr. S M Rashed Ul Islam, Associate Professor

Responsibilities and Experience:

• Detected malignant HNSCC samples via histopathology and screened for HPV using multiplex and nested PCR.
• Performed Sanger sequencing of the L1 viral gene and immunohistochemistry of host proteins for validation.
• Conducting molecular techniques to validate cancer-specific biomarkers retrieved through a multi-omics approach, including WGCNA, scRNA-seq, proteomics, and ChIP-seq.

Project:

• Genomic Exploration of HPV-Associated Head Neck Squamous Cell Carcinoma Occurrence in Bangladesh: An Integrative Histopathological Analysis and Molecular Profiling of HPV.
• Machine learning-driven identification and quantitative validation of cancer-specific biomarkers across multiple carcinomas.

Institutions in collaboration:

• Department of Surgical Oncology, Department of Otolaryngology–Head & Neck Surgery, and Department of Pathology, BMU; Department of Surgical Oncology, National Institute of Cancer Research & Hospital (NICRH); and BMB, SUST. More Info

GitHub:

• HNSCC-HPV, scRNA-seq

Supervisors: Dr. Tanvir Hossain, Assistant Professor; Preonath Chondrow Dev, Research Officer

Responsibilities and Experience:

• Developed omicML (GUI) to enable biologists to build biomarker algorithms from transcriptomic data.
• Utilizing deep neural network frameworks (e.g., VAEs, Transformers) for single cell segmentation and classification.
• Integrating patch-clamp electrophysiology, single-cell transcriptomics (Patch-seq, SMART-seq), and neuroimaging (fMRI) data to model neuronal excitability and cortical dysfunction in epilepsy.

Projects:

• omicML: an integrative tool of bioinformatics and machine learning algorithms to identify transcriptomic biomarkers
• Multi-modal computational decoding of neuronal cell types in epilepsy.

GitHub:

• omicML-server, omicML_RNA-seq, omicML-raw, Celltypes-Patchseq, Neuroimaging, Allen-Brain, Computational-Neuroscience

Supervisor: Dr. Ajit Ghosh, Associate Professor

Responsibilities and Experience:

• Leveraged random forest and XGboost to identify key salinity stress regulators in tomato, leading to the development of BioSalT.
• Extracted RNA from plant samples, quantified by Nanodrop, and performed PCR to confirm cDNA synthesis and primer specificity; visualized products on agarose gels.
• Validated candidate biomarkers by qRT-PCR using the ΔΔCt method to compute relative log2 fold changes.
• Profiled conserved domains and motifs of m6A regulators, built phylogenies (1000 bootstrap) in MEGA and visualized results in iTOL.

Projects:

• Genome-wide identification and characterization of m6A regulatory genes in Soybean: Insights into evolution, miRNA interactions, and stress responses.
• BioSalT (biomarkers of salinity stress in tomato): a multigene machine learning model for early salinity stress detection in Solanum lycopersicum.

GitHub:

• BioSalT, Genome-Wide

Supervisor: Dr. Tanvir Hossain, Assistant Professor

Responsibilities and Experience:

• Implemented a full RNA-seq workflow (bash) for gene quantification from GEO/SRA FASTQ files and performed differential expression analysis with edgeR, limma, and DESeq2.
• Built gene co-expression and network models to identify functional modules and hub genes enriched in biological pathways.
• Benchmarked 15 supervised learning models using PyCaret and validated biomarkers with the best-performing classifiers.
• Compiled a multi-source gene expression database and trained DeepFoldChange using MLPRegressor.

Projects:

• Identification of potential biomarkers for 2022 Mpox virus infection: a transcriptomic network analysis and machine learning approach.
• markerMPXV: upregulation of RRAD and building of a biomarker algorithm for Mpox virus infection in comparison with other viral pathologies. [Used as a case study for omicML]
• A deep fearning framework that emulates statistical models for differential gene expression analysis.

GitHub:

• Mpox-Project, Biomarker-Discovery, Bulk-RNA-seq, DeepFoldChange, DGE-analysis