Computational biologist working at the intersection of plant chromatin, proteomics, cancer genomics, and reproducible workflows.
I am a computational biologist and bioinformatician working on histone post-translational modifications (hPTMs) in plants, with a special focus on H3K79 methylation and acetylation in Arabidopsis thaliana. I combine quantitative mass spectrometry, reproducible computational pipelines, and long-read transcriptomics to understand how chromatin states change across development and stress.
Currently finishing my PhD at the University of Oviedo (Department of Organisms and Systems Biology), funded by an FPI fellowship (PRE2019-091395). My thesis combines four chapters: (1) software validation of EpiProfile_PLANTS with species-specific catalogs, (2) an ontogeny study of the Arabidopsis rosette across four developmental stages (YOUNG/BOT/FLOR/SEN), (3) re-analysis of public PRIDE datasets related to genotoxic stress, and (4) a synthesis of H3K79 behaviour across development and stress-related contexts.
A central piece of my work is EpiProfile_PLANTS—an extension of EpiProfile 2.0 tailored to plant histone proteomics. The ecosystem spans three interconnected repositories: the core MATLAB code with curated peptide catalogs for Arabidopsis, Marchantia polymorpha, and Chlamydomonas reinhardtii; a Docker + Snakemake workflow that handles PRIDE FTP download, msconvert, and MS1/MS2 extraction (220 raw files / 123 GB processed across PXD046034, PXD046788, PXD014739); and an interactive Dash/Plotly dashboard with 7 analysis tabs (heatmaps, PCA, volcano plots, PSM explorer, mass accuracy QC).
The data flows through a three-tier model: hDP (derivatized peptides) → hPF (peptideforms) → hPTM (site-level), with T1–T4 provenance tracking and a retention-time reference system for cross-run consistency.
Beyond proteomics, I develop K-CHOPORE—a 9-stage Snakemake + Docker pipeline for Oxford Nanopore direct RNA sequencing. It covers basecalling (Dorado/Guppy), isoform analysis (FLAIR/StringTie2), epitranscriptomic modification detection (ELIGOS2, m6Anet, xPore), and differential expression (DESeq2). Applied to a 2×2 factorial experiment in Arabidopsis (WT vs anac017-1), it yielded 20,958 isoforms and hundreds of DEGs. Version 3.0 adds five non-coding RNA modules including lncRNA discovery (FEELnc, CPC2), small RNA analysis (ShortStack, miRDeep-P2), and WGCNA co-expression networks.
Since November 2020 I have also been affiliated with the Computational Cancer Genomics Group at the Centro Nacional de Investigaciones Oncológicas (CNIO, Madrid). This experience spans NGS transcriptomics, lncRNA pipelines (VEp, FEELnc), and early-stage cancer genomics work, and shaped how I think about scalable, auditable bioinformatics. My 2017 publication on lncRNA clusterization in head and neck squamous carcinomas came from this line of research.
VIDIO (Vision-Integrated Diagnostic Imaging Orchestrator) is my multi-modal biomedical imaging platform, supporting retinal imaging (fundus, OCT), histopathology (OpenSlide whole-slide images), radiology (DICOM/NIfTI), and spatial transcriptomics (10x Visium, MERFISH). Built on Falcon WSGI, PyTorch/MONAI, OpenCV, with a PostgreSQL backend (16 tables) and TCGA/GDC integration.
I have designed and delivered courses at the Instituto Asturiano de Administración Pública (IAAP), the University of Oviedo, the City Council of Oviedo, FORMACAL, and ARTEAULA, covering Linux/WSL2/Docker, introductory Python and R/Bioconductor for omics, and hands-on projects linking code to real biological questions. Código Biológico is my growing outreach initiative to teach bioinformatics from scratch—step-by-step notebooks, recorded sessions, and reusable templates with real datasets, not toy examples.
Before and alongside the PhD, I have worked across multiple domains: GeoAI (geospatial / AI-driven analysis), ICM Lugo (research and healthcare context), and FSP-linked initiatives in regional public health. These experiences reinforced my philosophy: workflows should be practical, documented, and teachable.
Every analysis links back to explicit PXD accessions. Conversion from vendor files is containerised and deterministic. Intermediates (mzML, MS1/MS2) are standardised. Outputs are three-tier, audit-ready matrices. Documentation is citable. Code is open under GPL-family licenses. I follow FAIR principles and believe that if a result can’t be reproduced from raw data in a single command, it isn’t finished yet.
Thesis-linked projects, cancer bioinformatics, and reproducible reanalyses of public datasets.
Bottom-up histone proteomics across four rosette stages (YNG/BOT/FLOR/SEN), aiming to identify robust senescence-associated shifts in histone PTMs. The analysis emphasizes conservative QC, explicit normalization choices, and PTM summaries derived from peptideforms.
Reanalysis of multiple public datasets with the same processing contract (conversion, extraction, quantification, audit, statistics), to test robustness across instruments, batches and biological contexts.
Curating peptide catalogs/layouts for evolutionary comparisons across plant and algal systems, targeting conserved regions and reporting "core" PTM readouts in a consistent format.
H4 histone alignment: Arabidopsis, Marchantia, mouse & human sequences.
Bioinformatics work at the Centro Nacional de Investigaciones Oncológicas (CNIO), contributing to cancer genomics and transcriptomics analyses. Applying RNA-seq, differential expression, and multi-omics approaches to oncology research.
Independent validation of H3K79me3 changes across the four Arabidopsis rosette developmental stages (YNG, BOT, FLOR, SEN) by Western Blot, with synthetic K79me3 peptide control and H3 total loading reference.
Keen Comprehensive High-throughput Omics Pipeline Organizer. A 9-stage Snakemake + Docker pipeline for Oxford Nanopore direct RNA-seq, covering basecalling, isoform analysis, epitranscriptomic modification detection (m6A), and differential expression. Applied to an Arabidopsis thaliana 2×2 factorial design (WT vs anac017-1 mutant × control vs Antimycin A).
Biomedical image analysis platform for retinal, histology, radiology, and spatial transcriptomics pipelines. Combines computer vision with diagnostic imaging workflows.
Computational analysis of declassified documents from the 1981 Spanish coup d'état attempt. Named entity recognition (NER), thematic clustering, timeline reconstruction, and hypothesis-driven pattern discovery.
Curated meta-list of researchers, developers, and thought leaders across bioinformatics, AI, and computational biology.
Personal AI agent ecosystem: 5 departments, 13+ specialized agents, MCP integrations, and Claude-powered workflows for research automation and project management.
Plant-oriented EpiProfile 2.0 extension: quantification + QC + audit-ready outputs.
EpiProfile_PLANTS is a MATLAB-based suite for relative quantification of histone peptides and peptideforms from propionylation-based bottom-up proteomics. The key output layers are:
The plant extension focuses on curated species catalogs/layouts and on making the processing contract explicit: deterministic inputs, traceable intermediates (mzML to MS1/MS2), and QC artifacts that help interpret failures (e.g., retention time drift, empty windows, layout mismatches).
Keen Comprehensive High-throughput Omics Pipeline Organizer — Nanopore direct RNA-seq from raw signal to biological insight.
A 9-stage Snakemake + Docker pipeline for Oxford Nanopore direct RNA sequencing. The name is a playful nod to the Asturian cachopo—a layered sandwich dish—reflecting how the pipeline stacks analytical layers from raw signal to epitranscriptomic maps.
Currently applied to an Arabidopsis thaliana 2×2 factorial experiment (WT vs anac017-1 mutant × control vs Antimycin A), yielding 20,958 isoforms, 435 DEGs by genotype, and 266 DEGs by treatment (padj < 0.05, |log2FC| > 1).
| Condition | Wild Type (WT) | anac017-1 Mutant |
|---|---|---|
| Control | 3 replicates | 3 replicates |
| Antimycin A | 3 replicates | 1 replicate |
Unbalanced design handled via additive DESeq2 model (~ genotype + treatment).
The pipeline auto-switches between factorial and additive models based on replicate availability.
Reproducible workflows for RNA-seq, multi-omics, and data processing.
End-to-end RNA-seq processing: quality control (FastQC, MultiQC), alignment (STAR, HISAT2), quantification (featureCounts, Salmon), and differential expression analysis (DESeq2, edgeR).
FAIR-oriented pipeline design using Snakemake and containerized environments. Emphasis on version-controlled workflows, traceable intermediates, and deterministic outputs for bioinformatics analyses.
Multi-omics approaches combining proteomics, transcriptomics, and epigenomics data. Statistical integration and visualization strategies for complex biological datasets.
Visual diagrams mapping the experimental and computational pipelines across all thesis chapters.
Papers, preprints and contributed works.
Book chapter covering RNA sequencing platforms and bioinformatics tools for transcriptomics analysis. Part of Plant Transcriptomics and Epitranscriptomics (Springer, 2026).
Abstract: Identified distinct molecular clusters in HNSCC based on long non-coding RNA expression profiles. The clusters showed significant associations with clinical outcomes and known molecular subtypes.
Languages, bioinformatics tools, and infrastructure I work with daily.
Open-source tools, pipelines, courses, and project code.
Loading repositories...
Practical training for biologists: code, environments, and reproducible habits.
A hands-on course covering environment setup (WSL2, conda, VS Code), core Python patterns for data handling, and realistic examples from genomics/omics workflows.
Short courses on Linux for scientists, R/Bioconductor for omics data, and workflow management basics (Snakemake), with emphasis on reproducibility and documentation.
From R basics to advanced tidyverse workflows, ggplot2 visualization, statistical modeling, and Bioconductor packages for genomics and transcriptomics analysis.
Step-by-step tutorials covering the full RNA-seq analysis pipeline: quality control, alignment, quantification, differential expression, enrichment analysis, and publication-ready visualization.
A growing science-outreach initiative that teaches bioinformatics from scratch through real datasets and reproducible notebooks. Topics range from the genetic code (codon tables, translation) through the epigenetic code (histone PTMs, DNA methylation) to the chromatin code (nucleosome positioning, histone variants) and the computational code (R/Python scripts, Snakemake pipelines). Delivered as step-by-step Jupyter/R Markdown sessions with recorded walkthroughs and reusable templates.
Shared materials, protocols, cheat-sheets, and downloadable files.
Step-by-step protocols for histone extraction, propionylation, LC-MS/MS sample preparation, and data processing workflows.
Quick reference guides for R/Bioconductor, Python, Linux commands, Snakemake, Git workflows, and common bioinformatics one-liners.
Curated example datasets and sample outputs from EpiProfile_PLANTS and K-CHOPORE for testing and learning purposes.
Key figures from the PhD thesis — click any figure to view full size.
Bubble PCA biplot showing sample clustering by developmental stage. PC1 (17.3%) and PC2 (14.9%) capture separation between seedlings (YOUNG) and mature tissues (SEN, BOT, FLOR).
Violin-boxplots of all quantified histone PTMs faceted per modification. Each dot represents a biological replicate, colored by ontogeny stage (10H–SEN). Significance brackets from pairwise Wilcoxon tests.
All peptideform ratios (asin-sqrt) faceted by sequence and histone type (H3, H3.3, H2A, H4). Colored jitter reveals replicate distributions across five developmental stages of the Arabidopsis rosette.
Five-panel QC figure: (A) NA/zero/valid stacked bars, (B) group-level χ² tests, (C-D) top problematic peptides & modifications, (E) ratio distributions with Wilcoxon. 34 samples from the BGSR ontogeny dataset.
H4 Cross-Species Alignment
Compositional Distributions (ALR/CLR)
Pairs Scatter — Top EPI Types
H4 Peptide 4-17 Boxplot
Curated resources, tools, voices I follow, and references I use regularly.
People who inspire my thinking at the intersection of biology, data, and AI.
When I’m not at the bench or terminal, you’ll find me on the pitch.
Football is my reset button. I play for San Claudio in the Liga Asterov 2025/26 (CD Asterov, Asturias). Whether it’s scoring goals on the weekend pitch, watching a Champions League evening, or debating tactics over coffee, the beautiful game has been a constant in my life. The parallels between sports and science run deep: teamwork, strategy under pressure, and the beauty of an elegant solution.
| # | Player | Team | PJ | G | TA | TR |
|---|---|---|---|---|---|---|
| 1 | Jesús A. Abolí Fernández | Peluquerías Fran | 17 | 28 | 2 | 0 |
| 2 | Andrés Iglesias Lamuño | Metalasa | 15 | 23 | 2 | 0 |
| 3 | Rubén D. Arboleda Ospina | Cafeteros | 18 | 20 | 2 | 0 |
| 4 | Guillermo Vázquez Fernández | Ablaña Glassdrive | 17 | 17 | 1 | 0 |
| 5 | ⚽ Pelayo González de Lena | San Claudio | 16 | 17 | 3 | 0 |
| 6 | José M. Montes de Rueda | Manuel Cañedo | 15 | 16 | 3 | 1 |
| 7 | Andrés Lista | Peluquerías Fran | 12 | 16 | 2 | 0 |
| 8 | Alejandro Sierra Inguanzo | Olloniego City | 16 | 16 | 0 | 0 |
| 9 | Jhose M. Blanco Aranguren | Merendero El Cruce | 16 | 15 | 5 | 0 |
| 10 | Víctor Fernández Vázquez | Don Pelayo Dessert | 12 | 14 | 0 | 0 |
| Player | PJ | G | TA | TR |
|---|---|---|---|---|
| ⚽ Pelayo González de Lena Rodríguez | 16 | 17 | 3 | 0 |
| Juan Carlos Fernández Álvarez | 28 | 0 | 0 | 0 |
| Froilán Álvarez Fernández | 23 | 0 | 0 | 0 |
| José Ramón Fernández Mier | 19 | 0 | 0 | 0 |
| Adrián Gallego González | 15 | 2 | 1 | 0 |
| Santiago Marbán Lera | 15 | 0 | 0 | 0 |
| David Ríos Calle | 14 | 4 | 0 | 0 |
| Ramiro López Saiz | 13 | 0 | 1 | 0 |
| Alejandro Pesquera Bengoa | 13 | 0 | 0 | 0 |
| Pablo Fernández García | 13 | 0 | 1 | 0 |
| Francisco García Gijón | 12 | 3 | 4 | 0 |
| José Ignacio García García | 12 | 0 | 1 | 0 |
| Francisco José Álvarez Álvarez | 11 | 0 | 0 | 0 |
| Mariano García Álvarez | 9 | 1 | 0 | 0 |
| Álvaro Suárez Martín | 9 | 1 | 0 | 0 |
| Daniel Sánchez de Dios | 8 | 0 | 0 | 0 |
| José Diego Ramos Fernández | 8 | 1 | 0 | 0 |
| Ignacio Rosa Cuesta | 8 | 1 | 0 | 0 |
| Eduardo Rivero Fernández | 7 | 1 | 2 | 0 |
| Iván Bautista Caicedo | 7 | 0 | 0 | 0 |
| Andrés Álvarez Álvarez | 5 | 0 | 1 | 0 |
| Daniel Alestar Chirosca | 5 | 1 | 0 | 0 |
| Roberto D. Bengoa Pesquera | 5 | 1 | 0 | 0 |
| Iván González Bada | 4 | 0 | 0 | 0 |
| Igor Cabrera Eguiluz | 3 | 0 | 1 | 0 |
| Manuel Vázquez Riva | 3 | 0 | 1 | 0 |
| Javier Martínez Aller | 3 | 0 | 0 | 0 |
| Borja F. Barbón Velasco | 2 | 0 | 1 | 0 |
| David Martínez Diego | 2 | 0 | 0 | 0 |
| Julio B. Cortés Hernando | 2 | 0 | 0 | 0 |
| Pablo Fernández Zapico | 2 | 1 | 0 | 0 |
| Miguel Díaz Arias | 2 | 0 | 0 | 0 |
| Jorge Fernández Cachero | 1 | 0 | 0 | 0 |
| Luis A. Colunga Suárez | 1 | 0 | 0 | 0 |
| Oliver González González | 1 | 0 | 0 | 0 |
| Constantin Gabriel Vreja | 1 | 0 | 0 | 0 |
| Juan Camilo Rodríguez Lasso | 1 | 0 | 0 | 0 |
| Pablo Blanco Menéndez | 0 | 0 | 0 | 0 |
All stats from the official CD Asterov platform. Last updated: 25 Feb 2026 · Next match: Jornada 20 vs CF Cuelebre (1 Mar)
Best ways to reuse the code, follow updates, or get in touch.
Based at the Department of Organisms and Systems Biology (University of Oviedo). Open to discussions around plant chromatin, cancer bioinformatics, histone PTM proteomics, and reproducible workflows.
This website is updated as thesis work, software releases and dataset reanalyses mature. Last update: March 2026.