Research

Below are selected projects and publications. Please refer to my CV or Google Scholar for a complete list.

DrugMechDB:  A curated Database of Drug Mechanisms

Drug repositioning methods based on biomedical knowledge graphs typically offer useful supporting biological evidence. This evidence is based on reasoning chains or subgraphs that connect a drug to a disease prediction. However, there are no databases of drug mechanisms that can be used to train and evaluate such methods. Here, we introduce the Drug Mechanism Database (DrugMechDB), a manually curated database that describes drug mechanisms as paths through a knowledge graph. DrugMechDB integrates a diverse range of authoritative free-text resources to describe 4,583 drug indications with 32,249 relationships, representing 14 major biological scales. DrugMechDB can be employed as a benchmark dataset for assessing computational drug repositioning models or as a valuable resource for training such models.

Scientific Data, Github Repository, Data

Effects of autism spectrum disorder (ASD)-associated  variants on C. elegans development

We investigated how ASD-associated mutations in the daf-18/PTEN gene affect development in C. elegans. We focused on five different mutations and tested their effects on the worm's ability to enter the dauer state, which is a dormant developmental stage that occurs under harsh conditions. After developing an assay using pheromones and limited food to induce dauer formation, we discovered that two of our mutations (H138R and T176I) significantly impaired the worms' ability to enter dauer state, while the other three mutations (D66E, L115V, and H168Q) didn't show much effect. Most of our tested mutations were in the phosphatase domain of the DAF-18 protein, with one particularly interesting mutation (H138R) affecting the protein's active site. We believe this explains why it had such a strong effect on dauer formation. Through this work, we've shown that C. elegans can serve as an effective model for studying ASD genetics, and we've contributed to the broader understanding of how these genetic variations might affect development and function.

MicroPublication Biology

KRAS, NRAS, and BRAF Mutations in Mexican Metastatic Colorectal Cancer: Clinical Correlations and Predictive Modeling

We investigated the prevalence of mutations in KRAS, NRAS, and BRAF genes among 500 Mexican patients with metastatic colorectal cancer (mCRC) and analyzed how these mutations relate to clinical and pathological features. We found that 52% of patients had clinically relevant mutations, with KRAS mutations being most common (86% of mutated cases), followed by NRAS (7%) and BRAF (6%). When looking at correlations with tumor characteristics, we discovered that tumors located in the proximal colon were significantly more likely to have KRAS and BRAF mutations.

PloS one

Sex specific function of eputhelial STAT 3 signaling in KRAS  mutant lung cancer 

We investigated how STAT3 affects lung cancer development in mice with K-ras mutations and found surprising sex differences. When we deleted STAT3 in lung cells, female mice developed fewer tumors while male mice showed increased tumor growth. We discovered this was because females developed stronger anti-tumor immune responses, while males showed more tumor-promoting inflammation. In males, blocking the inflammatory protein IL-6 reduced tumor growth. In females, we found that estrogen played a protective role - when we blocked estrogen signaling, tumors grew more aggressively. This findings suggest that male and female lung cancer patients might benefit from different treatment approaches, with males potentially benefiting from anti-inflammatory treatments while preserving estrogen signaling could be important for females.

Nature Communications