Natalia De Oliveira Rosa

Atlantis integrates structural and functional residue-level data for the human proteome, enabling complex queries and analysis of protein interactions and mutations.

This study identifies a LRRK2-regulatory network through proximity proteomics and structural analysis, showing its link to centrosomes and how its conformation and activity state influence interactions.

This study predicts 3D structures of human GPCR-G protein complexes using AlphaFold3 and uses machine learning to predict coupling potency, revealing new signaling mechanisms and coupling preferences across GPCR subtypes.

The paper identifies Gs pathway inactivation as an SMO-independent Hedgehog signaling driver in BCC and shows adenosine 2B receptor activation can counteract SMO-driven oncogenesis.

The paper develops interpretable models to predict cancer drug sensitivity using transcriptomics data and enhances predictions with large language models, validated on clinical samples.

The paper analyzes GPCR signaling dysregulation in cancer transcriptomics data to identify therapeutic targets and survival-associated signaling axes.