XenoSignal: Investigating Intra- and Inter-Species Ligand-Receptor Interactions Using AlphaFold3
Abdelkareem, A. O.; Hall, C. F.; Marutharaj, A.; Narta, K.; Verhey, T. B.; Morrissy, A. S.
Show abstract
Xenograft models, in which human tumors are implanted into mouse hosts, offer a unique platform to study tumor-microenvironment (TME) interactions. Ligand-receptor (LR) interactions are central to intercellular communication in this context, but most inference tools rely on single-species databases and assume conserved function across orthologues. To test whether this assumption holds given sequence and structural divergence, we developed XenoSignal, a computational framework that integrates AlphaFold3 structural predictions to assess the plausibility of intraspecies and interspecies LR interactions. The pipeline predicts whether human-mouse orthologues retain or lose their functional interaction, enabling construction of a high-confidence cross-species LR resource. XenoSignal facilitates the interpretation of tumor-TME crosstalk in xenografts and prioritizes LR pairs for experimental validation. This resource enhances our ability to dissect interspecies signaling, uncover mechanisms of immune evasion, and identify candidate therapeutic targets in translational cancer models.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Evaluation of AlphaFold Antibody-Antigen Modeling with Implications for Improving Predictive Accuracy 96%
- COLLAPSE: A representation learning framework for identification and characterization of protein structural sites 96%
- Neural Network-Derived Potts Models for Structure-Based Protein Design using Backbone Atomic Coordinates and Tertiary Motifs 96%
Similar papers in this journal
Similar papers in this journal
Similar papers in this journal
- BindPred: A Framework for Predicting Protein-Protein Binding Affinity from Language Model Embeddings 95%
- Deep Local Analysis evaluates protein docking conformations with locally oriented cubes 94%
- Deep Local Analysis deconstructs protein-protein interfaces and accurately estimates binding affinity changes upon mutation 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.