Sandra Castillo

Comp2GPR improves gene-protein-reaction rules by integrating curated protein complex data with sequence-based validation, enhancing metabolic model accuracy.

ActSeekN improves enzyme function annotation by using structural motifs from AlphaFold and UniProt data, outperforming existing methods in yeast, human, and Trichoderma reesei proteomes.

This study uses a conditional variational autoencoder to design new-to-nature PHA synthases, which are validated for activity in yeast and show potential for expanding PHA material properties.

ActSeek searches the Alphafold database for proteins with active sites similar to a seed protein, identifying enzymes for biodegradable plastics and drug off-targets.

The paper uses genome-scale metabolic models to identify metabolic differences between non-Saccharomyces yeasts, finding that Complex I in the electron transport chain distinguishes Metschnikowia pulcherrima from other species.

The paper develops a deep learning method to reconstruct compartmentalized genome-scale metabolic models for over 800 fungi, improving predictions of metabolic phenotypes and phylogenetic clustering.

The paper introduces PMFA, a method combining PCA and stoichiometric flux analysis for metabolic flux mode analysis, showing its versatility through experiments.