Dr. Jeferyd Yepes
Jeferyd is a bioinformatician and computational biologist specialising in metagenomics, microbial genomics, reproducible bioinformatics workflows, and artificial intelligence for protein function prediction. His research combines large-scale metagenomic analysis with modern machine learning approaches to understand complex microbial communities and their functional potential.
Jeferyd completed his PhD in Bioinformatics at the University of Fribourg and the Swiss Institute of Bioinformatics (SIB), supported by a Swiss Government Excellence Scholarship. His doctoral research focused on understanding rice straw degradation using metagenomics, with particular emphasis on reconstructing Metagenome-Assembled Genomes (MAGs) to identify microorganisms with lignocellulose-degrading potential.
Alongside his metagenomics research, Jeferyd works with protein language models (pLMs), incorporating structural information and fine-tuning machine learning models to improve automated protein function prediction from microbiome data. His research has received international recognition, including selection as an SIB Remarkable Output in 2024 and the Best Poster in Bioinformatics and Systems Biology at the Life Sciences Switzerland (LS2) Annual Meeting 2025.
Education & Career
- PhD in Bioinformatics, University of Fribourg, Switzerland (2026)
- Bioinformatics Researcher, Swiss Institute of Bioinformatics (SIB), Switzerland
- Master’s degree in Engineering, University of Antioquia, Colombia
- Former Laboratory Analyst at Iluma Alliance and the University of Antioquia, Colombia
- Recipient of a Swiss Government Excellence Scholarship for doctoral research
Research Focus
Jeferyd’s research centres on developing and applying computational approaches for analysing complex microbial communities and extracting functional information from large-scale metagenomic datasets. His research interests include:
- Metagenomics and microbial community analysis
- Metagenome-Assembled Genome (MAG) reconstruction
- Microbial genome recovery, quality assessment, and taxonomic annotation
- Bioinformatics pipeline development and benchmarking
- Nextflow and reproducible computational workflows
- Protein language models and machine learning
- Protein function prediction from microbiome data
- Lignocellulose degradation and microbial biotechnology
Main Achievements
During his PhD research, Jeferyd:
- Developed a suite of bioinformatics tools supporting metagenomics pipeline selection, benchmarking, development, and reproducible analysis
- Performed comprehensive metagenomic analyses, including MAG reconstruction, identifying microorganisms with substantial lignocellulose-degrading potential during rice straw decomposition
- Applied and fine-tuned protein language models to improve prediction of lignocellulose-degrading enzyme functions, outperforming traditional sequence-based approaches
- Developed resources designed to make sophisticated metagenomics workflows more accessible and reproducible for researchers
Awards
- SIB Remarkable Output 2024 — Swiss Institute of Bioinformatics
- Best Poster in Bioinformatics and Systems Biology — Life Sciences Switzerland (LS2) Annual Meeting 2025
- Swiss Government Excellence Scholarship (2022) — Federal Commission for Scholarships for Foreign Students
Teaching & Training
Jeferyd has taught bioinformatics and computational biology through the Swiss Institute of Bioinformatics, with particular expertise in metagenomics, reproducible workflows, Nextflow, and scientific data visualisation. Previous courses include:
- Nextflow in Action: Build Smarter, Faster, Reproducible Pipelines (2026)
- Introduction to Metagenomics Data Analysis of Microbial Communities (2024, 2025, 2026)
- Interactive Visualization with Python Fribourg (2024, 2025)
- Practical training in reproducible bioinformatics workflows and computational analysis
Tools Developed
- 2Pipe — A decision-support tool designed to help researchers select appropriate pipelines for Metagenome-Assembled Genome reconstruction
- MAGFlow and BIgMAG integration — Tools for visualising metagenome quality metrics and taxonomic annotations
- TaxoFlow — A step-by-step framework and tutorial for constructing Nextflow pipelines for metagenomic taxonomic classification