About Me
Hi, my name is Esteban. I’m PhD in science research with expertise in bioinformatics and machine learning applied to immunology, genomics, proteomics, and lipidomics. Experienced in developing predictive tools for health research, analyzing large-scale biological data, and working with programming languages like Python, R and Bash. My strong background in both bioinformatics and lab techniques helps me to bridging multidisciplinary teams to solve complex problems.
I’m currently interested in improving my skills in machine learning and developing Python applications for biological and health research.
Repositories
Machine learning and rheumatoid arthritis treatments
github.com/eagomezc/Machine-Learning-and-RA-treatmentThis repository contains the script used to study blood pro-resolving mediators as biomarkers to predict the response of DMARD treatment in rheumatoid arthritis patients.
Genome association analysis and rheumatoid arthritis
github.com/eagomezc/CG-association-analysis-in-SPM-related-genesThis repository contains the scripts used to study the genetic associations between genetic variant in SPM-related genes and rheumatoid arthritis
This repository contains the script used for the developing of a web-based application for the identification of retrotransposons in human genomes.
Enriched marine oils supplements effect in human health
github.com/eagomezc/2019_DGE_Correlation_Oil_supplements_studyThis repository contains the scripts used to run the differential gene expression, GO enrichment and correlation analysis for the paper: “Enriched marine oils supplements increase peripheral blood SPM concentrations and reprogram host immune responses: A randomized double-blind placebo-controlled study”
Experience
Innovative thinker, bioinformatics solutions, machine learning, wet-lab training
Machine learning: I worked with machine learning methodologies: support vector machine, bayesian classifiers, and decision trees (R randomForest and caret, scikit-learn in Python) to predict the response of different treatments in patients with rheumatoid arthritis. The machine learning models were patented.
Network analysis: I created small tools for the design of biosynthetic networks (Cytoscape, R Shiny) to identify different responses to treatments and up or down-regulation of immunological pathways.
Creative problem solving: The Lipid Mediator Unit is a multidisciplinary team that includes immunologists, chemists and biologists who rely on me to help them with multivariate analysis, statistics, gene expression analysis, data visualization, etc.
Lab skills: As bioinformaticians, it’s important to know where the data comes from; for that reason, I received training in mass spectrometry, flow cytometry, cell isolation and cell culture assays.
Prospection and Design of Biomolecules group, UNAL
Researcher
2015 - 2017
ciencias.medellin.unal.edu.co/gruposdeinvestigacion/prospeccionydisenobiomoleculas/
Fast learner, Proactive, Team work, bioinformatics and databases tools
Fast learner: From scratch, I curated and designed a database of antimicrobial peptides (InverPep). I had to learn strategies to organize, curate and build the database. In addition to this, I developed a friendly user interface.
Proactive: The identification of bioactive peptides in proteomes was done manually. I developed a tool (PepMultiFinder) that based on physicochemical properties automatically identifies antimicrobial peptides.
Education
Queen Mary University of London
PhD Biochemical Pharmacology
2019 - 2024
Doctor of Philosophy (Focused in Bioinformatics).
PhD thesis: Exploring the potential of specialized pro-resolving mediator biology in understanding the development and progression of inflammatory diseases using bioinformatics.
Technicals system used: Univariate and multivariate analysis (PCA, PLS-DA, correlation analysis), machine learning (Bayesian classifiers, Random forest, LASSO, SVM), HPC (LSF), RNA (Bulk and scRNA-Seq) data analysis (multiQC, STAR reads mapping, FeatureCount alignment, R Seurat), Differential Gene Expression analysis (EdgeR), Genome and Pheno-wide association analysis (UK Biobank, PLINK, Python, Linux, Bash, Anaconda), GO enrichment, Network and pathway analysis (Cytoscape, R Shiny, STRING, KEGG, Reactome), flow cytometry and liquid chromatography-mass spectrometry (lipidomics).
Supervision experience: I mentored three master’s students who were doing their master’s dissertations in bioinformatics at QMUL.
Queen Mary University of London
MSc. Bioinformatics
2017 - 2018
Master in Science with an average of 82.37/100 equivalent to a Distinction.
Master dissertation: Searching for signatures of selection in a highly social insect – A population-genomics analysis. Ecology, evolution, genome-wide screening and genome assembly.
Developed Software - Database: Retrotransposon: a web-based database for the collection, identification and analysis of retrotransposons (SQLite, Flask, HTML, R).
Relevant modules: Genome Bioinformatics, Coding for Scientifics, Post-Genomics Bioinformatics and Statistics.
Technicals system used: Genome assembly programs (SOAPdenovo and Masurca), Python (flow control, functions, objects, biopython), R and RStudio, machine learning, Proteomics tools (MetaboAnalysti, Expasy, I-Tasser).