How do proteins evolve?
We study the origin and diversification of protein families, from the evolution of signalling systems to the emergence of functional novelty and convergence.
We study how proteins evolve, acquire new functions and adapt their structures. Our work brings together evolutionary biology, computational methods and protein language models to explore the diversity of the protein world.
Research is collaborative. Behind each paper or resource are people, questions, discussions, and experiments.
We start with biology. Computational approaches help us investigate questions that experiments and conventional sequence comparisons cannot easily resolve.
We study the origin and diversification of protein families, from the evolution of signalling systems to the emergence of functional novelty and convergence.
Protein multifunctionality, structural plasticity and fold switching challenge simple relationships between sequence, structure and function.
We develop computational approaches to explore poorly annotated proteomes, including the use of protein language models and embedding-based functional annotation.
Our software and analyses connect sequences, structures, evolutionary histories and biological function.
Explore our GitHub ↗
Embedding-based functional annotation of proteins, designed to bring protein language models to the study of non-model and poorly characterised proteomes.
Research outputs / 2012—2026
Bastolla U, Aurell E, Cocco S, Liberles DA, Micheletti C, Orellana L, Rojas A, Talavera D & Valencia A. Journal of Molecular Evolution (2026). Published 4 November 2026.
Read published article ↗Protein Language Model Embeddings Recover Evolutionary and Functional Organization Across the Ras Superfamily.
PublicationFANTASIA suite: a reproducible and configurable framework for embedding-based functional annotation of proteins.
PublicationRethinking molecular evolution through protein language model embeddings.
PublicationFANTASIA leverages language models to decode the functional dark proteome across the animal tree of life.
CBBIO/FANTASIA, CBBIO/FANTASIA-Lite
PublicationDecoding proteome functional information in model organisms using protein language models.
PublicationThe multivalency of the glucocorticoid receptor ligand-binding domain explains its manifold physiological activities.
PublicationOrigin and Evolution of Polycyclic Triterpene Synthesis.
PublicationNovel miRNA-mRNA interactions conserved in essential cancer pathways.
PublicationA non-canonical mismatch repair pathway in prokaryotes.
PublicationEmergence and evolutionary analysis of the human DDR network: implications in comparative genomics and downstream analyses.
PublicationThe Ras protein superfamily: Evolutionary tree and role of conserved amino acids.
PublicationOur group
Principal Investigator
Research Scientist, CSIC · Director, Evolutionary Dynamics, CABD
Associate Investigator
Tenured CSIC Scientist · CABD Bioinformatics Unit HeadPhD Student
Biology · computational methods
PhD Student
Biology · computational methods
PhD Student
xAI for multifunction prediction
Programmer Technician
Technical support, programming, and Machine Learning
TFM Student
JAE-ICU
CSIC Technician
Machine Learning, Data Science and Structural BioinformaticsFormer Members
Students
Visitors
Collaborators
Group leadership
Ana M. Rojas Mendoza is a CSIC Research Scientist and, since 2026, Director of the Department of Evolutionary Dynamics at CABD.
Trained in Biological Sciences in Madrid (Spain) and later specialized through predoctoral and postdoctoral research periods in the United States and Spain, she has built a career connecting molecular biology, evolution, structural bioinformatics, and computational methods. She established her independent group in 2009 in Barcelona, and later moved to Seville first at IBiS and later CABD, where she is currently leading the group.
Her scientific trajectory spans protein family evolution, protein signaling, omics integration, and methodological development, with a clear transition in recent years toward artificial intelligence and protein language models. Current work focuses on how sequence, structure, and evolution shape protein function, and on making functional annotation more scalable and more biologically meaningful.
She combines research leadership with teaching, mentoring, scientific evaluation, and international community work, including active roles in Function COSI, SEBiBC, and CSIC research networks.
Training
Ongoing doctoral supervision, master’s projects, and visiting researchers span computational biology, structural bioinformatics, and AI-based annotation.
Ana M. Rojas teaches in the following master’s programs at the University of Seville:
She is also an external lecturer in the doctoral program Ingeniería, Ciencia de Datos y Bioinformática.
Location
Andalusian Center for Developmental Biology (CABD/CSIC-UPO-JA)
Avda. Rectora Rosario Valpuesta 1
41089 Dos Hermanas, Sevilla
Contact
For research collaborations, student enquiries, or speaking invitations, contact the lab directly.
CABD Phone: +34 954977911
UPO internal: 67911
Security: +34 954349200
Fax: +34 954349376
Head profile