NiccolĂ˛ Pancino (University of Florence)

###### Dec 04, 2019 – 11:00 AM

###### DIISM, Artificial Intelligence laboratory (room 201), Siena SI

##### Description

Binding site identification allows to determine the functionality and the quaternary structure of proteinâ€“protein complexes. Various approaches to this problem have been proposed without reaching a viable solution. Graph theory is a very helpful instrument in this task. Representing the interacting peptides as graphs, a correspondence graph describing their interaction can be built. As previously demonstrated, finding the maximum clique in the correspondence graph allows to identify the secondary structure elements belonging to the interaction site. Although the maximum clique problem is NP-complete, Graph Neural Networks make for an approximation tool that can solve the problem in affordable time, having both the computational power of a deep neural network and the capability of processing graph structured inputs natively.

In this seminar, after a briefly introduction to Proteomics and Graph Theory, the data preprocessing will be described, from the graph representation of protein molecules to the construction of the correspondence graphs composing the dataset for the GNN training. Finally, experimental results will be reported, outlining possible future perspectives.