3. Data, Science & AI

Handles creation, manipulation, and conversion of bedGraph files for genomic data visualization, including normalization and bigWig conversion for ChIP-seq, ATAC-seq, and RNA-seq.

Performs normalization, batch correction, feature alignment, and missing value handling for multi-omics data integration.

Performs species-level taxonomic profiling using MetaPhlAn 4's clade-specific markers for accurate microbial community analysis with minimal computational resources.

Handles parsing, querying, and conversion of GTF and GFF3 genomic annotation files to extract gene, transcript, and exon coordinates using bioinformatics tools.

Builds reproducible bioinformatics pipelines using Snakemake with dependency tracking for multi-step data analysis on HPC clusters.

Performs DIA proteomics analysis using DIA-NN for library-free and library-based workflows to enable deep proteome profiling.

Executes Gene Set Enrichment Analysis (GSEA) via clusterProfiler to detect coordinated gene expression changes from ranked gene lists without arbitrary significance thresholds.

Processes VCF genomic variant files via bcftools for merging, sorting, intersecting, and subsetting to enable variant analysis.

Analyzes pooled CRISPR-Cas9 screen data using MAGeCK for gene ranking, pathway analysis, and identification of essential genes or drug targets.

Predicts metagenome functional content from 16S rRNA data using PICRUSt2, inferring KEGG, MetaCyc, and EC abundances without shotgun sequencing.

Performs quality control on genomic phasing and imputation results, filtering by INFO scores and validating accuracy for GWAS-ready data preparation.

Creates publication-ready multi-panel figures from multiple plots using R packages patchwork, cowplot, and gridExtra with shared legends and annotations.

Filters biological sequences by length, GC content, ID, or patterns using Biopython for data subset selection.

Aligns Oxford Nanopore and PacBio long reads to reference genomes using minimap2 for variant calling, SV detection, and coverage analysis.

Deep learning-based variant calling for SNPs and small indels from long-read genomic data (ONT/PacBio) using Clair3, ensuring high accuracy for clinical and research applications.

End-to-end pipeline for structural variant detection in long-read genomic data using minimap2 alignment and Sniffles/cuteSV SV calling.

Efficiently downloads large biological datasets from NCBI using batching and rate limiting for bulk sequence and query results.

Automates batch processing of biological sequence files (merging, splitting, directory operations) using Biopython.

Performs advanced sequence similarity searches for bioinformatics, identifying orthologs and distant homologs using PSI-BLAST, HMMER, and reciprocal best hit methods.

Profiles metagenomic functional potential by analyzing pathway abundances, gene family counts, and functional annotations using HUMAnN3 and similar tools.

Generates reproducible scientific documents, presentations, and websites using Quarto with R, Python, Julia, and Observable JS support.

Handles Hi-C genomic data by loading, converting, and manipulating contact matrices in cooler format, supporting .cool, .mcool, and .hic file types.

Analyzes spatial relationships and cell-cell interactions in imaging mass cytometry (IMC) data using spatial statistics and neighbor graph methods.

Provides a command-line workflow for 16S/ITS microbiome amplicon analysis with reproducible provenance tracking, serving as an alternative to R-based tools.