I focus on bioinformatics, transcriptomics, and computational genomics, with an interest in reproducible workflows and evidence-based evaluation of genomic analyses.
| Area | Demonstrated skills |
|---|---|
| Analysis | RNA-seq, differential expression, germline variant analysis, GIAB benchmark interpretation, homology-based functional annotation |
| Programming | R, Python, Bash |
| Workflow tools | Nextflow DSL2, Conda, Linux/WSL |
| Bioinformatics tools | DESeq2, HISAT2, featureCounts, DIAMOND, BWA-MEM, GATK, Picard, SAMtools, BCFtools, FastQC |
A 13-process Nextflow DSL2 workflow from paired-end FASTQ reads to normalized, hard-filtered PASS germline variants, validated on HG002 (SRR2962669).
- Uses FastQC, BWA-MEM, Picard, GATK, samtools, and bcftools for read QC, alignment, coverage assessment, BQSR, variant calling, and filtering.
- Independent GIAB benchmarking achieved 97.38% precision, 95.83% sensitivity, and 96.60% F1 within the specified 47.41 Mb autosomal WES benchmark territory.
- GIAB benchmarking, VEP annotation, and exploratory variant prioritization were separate downstream analyses, not integrated steps of the Nextflow workflow.
Benchmark methods and limitations · Execution provenance · Annotation and prioritization
This project demonstrates workflow development, germline variant analysis, reproducibility, and benchmark interpretation. The reported results are specific to the defined HG002 evaluation territory and do not constitute clinical validation.
A transcriptomic analysis of licorice leaves across control, 6 h, 12 h, and 24 h insect-infestation groups, using 12 paired-end samples with three biological replicates per group.
- Uses a per-sample Bash workflow for read processing, HISAT2 alignment, and featureCounts gene counting, followed by DESeq2 treatment-versus-control comparisons.
- Uses separate downstream DIAMOND best-hit mapping to Arabidopsis thaliana and GO/KEGG over-representation analyses to identify candidate functional associations.
- Includes sample-to-SRA accession mapping, analysis scripts, expression tables, PCA and other visualizations, and documentation of reproducibility limitations.
Study design and workflow · Results and interpretation · Analysis outputs
This project demonstrates RNA-seq processing, differential expression analysis, annotation, and interpretation of enrichment results. The computational findings support hypotheses for follow-up; they do not establish pathway activity, metabolite accumulation, or causal defense mechanisms.
My project work spans human germline variant analysis and plant transcriptomics. I focus on traceable analyses, reproducible workflows, and careful interpretation of computational findings.
For technical questions about a project, use its GitHub issue tracker: Germline WES · Plant RNA-seq.