Adrián Jinich

ProtoPure automates the extraction and analysis of protein purification protocols from literature, revealing trends in recombinant expression practices and diverse downstream purification strategies.

This paper designs de novo minibinders that inhibit blood-stage malaria replication with nanomolar potency by targeting the RH5-basigin interaction.

The paper identifies alkamines as a previously hidden class of metabolites formed via reductive amination, revealing significant gaps in metabolomics and drug metabolism understanding.

This paper shows that protein language models can detect local structural similarities from sequence data without 3D models, identifying aligned regions through cosine similarity of residue embeddings.

The paper introduces microbiomeMASST, a metadata network graph linking metabolites across datasets, enabling cross-study analysis of microbial metabolites and their biological context.

TubercuProbe prioritizes compound-protein interactions in Mycobacterium tuberculosis using a cross-attention graph-sequence model trained on human chemoproteomic data.

The paper introduces PLVis, a framework using protein language model embeddings to visualize and explore protein homology, validated through case studies on Mycobacterium and Plasmodium proteomes.

The paper uses synthetic multiplexing to annotate uncharacterized metabolites, identifying ibuprofen-carnitine as a metabolite linked to muscle toxicity.

The paper identifies genes in Mycobacterium tuberculosis essential for intra-alveolar survival after airborne transmission, revealing stage-specific roles in transmission.

This study identifies a periplasmic protein complex in Mtb that supports arabinofuranosyltransferase activity and contributes to intrinsic drug resistance.

This study identifies a candidate transmission survival genome of Mycobacterium tuberculosis by exposing it to model aerosol fluid and analyzing gene contributions to survival during transmission.

The paper uses transposon mutagenesis and QTL mapping to identify host-pathogen genetic interactions affecting Mtb fitness in immunodivergent mice.

The paper introduces PLUMAPs, a tool that visualizes protein similarity using protein language models, dimensionality reduction, and topic modeling, showing improved clustering and interpretability compared to traditional sequence similarity networks.

The paper predicts enzyme substrate selectivity using space-filling curve-based structural representations of proteins, achieving high accuracy with gradient-boosted classifiers.

The paper compares the thermodynamic properties of biological and non-biological compounds in redox chemical space, finding metabolites have lower Gibbs energies and specific functional group enrichments.