GMM Desktop App
A complete, tested desktop workflow for importing, aligning, analysing, visualising and exporting landmark-based shape data.
Read case studyBiological questions · computational methods
Guillermo Bravo Morante Scientific Data Analyst · Research Technician · Software Builder
I combine scientific rigour, 3D data, statistical analysis and software development to turn complex problems into clear and usable results.
What I work with
Four connected capabilities
The methods vary, but the underlying work is consistent: structure difficult data, make the process reliable and communicate what the evidence can—and cannot—support.
01
Data analysis & statistical modelling
From cleaning and exploration to validation, interpretation and reproducible reporting.
↗ 02Scientific visualisation & communication
Clear plots, publication graphics and explanations for students, collaborators and decision-makers.
↗ 033D scanning & quantitative morphology
Thousands of surface-scanned skeletal elements, Artec workflows, landmark data, morphometrics and mesh quality control.
↗ 04Software, automation & reproducible workflows
Practical applications and repeatable pipelines shaped around real analytical problems.
↗Selected projects
From analysis to usable tools
Selected systems built around real use: automated 3D landmarking, scientific analysis, statistical experimentation and practical learning workflows.
Automated 3D Landmarking Pipeline
An end-to-end desktop pipeline that turns CT studies and surface scans into analysis-ready anatomical landmarks with uncertainty-led human review.
Read case studyMulti-Species Pelvis Morph Explorer
An interactive 3D comparison that morphs one shared pelvic surface among human, chimpanzee, bonobo, gorilla and orangutan forms.
Read case studyPolymarket Lab — Statistical Strategy Test-Bench
A self-hosted paper-trading laboratory that tests prediction-market strategies against independent virtual portfolios and a no-action control.
Read case studyCurrent analytical work
Short cases from real research problems
Anonymised problem–solution notes from methodological work with students: what threatened the analysis, what I changed and why the decision mattered.
When observer differences mimic a biological signal
I redesigned the tests around the conditional question—group effect after accounting for observer—and controlled the resulting family of comparisons with false-discovery-rate correction. A landmark-participation table then localized where any remaining distance effect was concentrated.
Adding curved anatomy without assuming it improves the answer
I built the curve, redistribution and sliding workflow, then ran three parallel analyses: fixed landmarks only, semilandmarks only and both together. The design made the influence of the new representation visible instead of assuming that more points were automatically better.
Preventing one anatomical block from dominating another
I used a Multiple Factor Analysis-style balancing step before classification, then applied shrinkage discriminant analysis with stratified cross-validation and balanced accuracy. The held-out target remained completely outside training until the final classification.
The career bridge
Scientific experience is data experience.
- 01
Messy data
Real biological datasets are incomplete, variable and shaped by how they were collected.
- 02
Reproducible decisions
Analysis needs explicit steps, validation and a record that another person can inspect.
- 03
Uncertainty, made visible
A useful result communicates its limitations as clearly as its main pattern.
- 04
Tools around repeated work
Recurring analytical friction is often a product opportunity: automate the routine and keep judgment visible.
Selected evidence
Work that can be inspected
Peer-reviewed research is one form of evidence. Reusable workflows, teaching and clear documentation are others.
View the publication catalogue- 2025
- 2025
Veracidad vs. engaño: detección de indicios verbo-corporales en el discurso
Revista Criminalidad
Open DOI ↗ - 2024
Convergent evolution in Afrotheria and non-afrotherians demonstrates high evolvability of the mammalian inner ear
Nature Communications
Open DOI ↗
Current direction
Building a technical career that keeps scientific rigour at its core.
I’m expanding my work in data engineering, application development and automation while continuing to solve research problems. The next step is not away from science; it is toward roles where rigorous analysis and useful software meet.