Alessio Del Conte

Ritratto di Alessio Del Conte

Curriculum
Neuroscience, Technology and Society, XXXVII series
Supervisor
s
Silvio Tosatto
Contact
alessio.delconte@studenti.unipd.it



Project description
My research projects aim to study low complexity binding sites of proteins, using machine learning techniques to determine where these sites are located and possibly their function. For decades biologists and biophysicists have studied the three dimensional structure of proteins because it is generally assumed that the function of a protein is closely linked to its three-dimensional structure. Despite this, it has been discovered that a large proportion of gene sequences appear to code not for structured, completely folded, globular proteins, but for long stretches of amino acids that are likely to be either unfolded in solution or adopt non-globular structures of unknown conformation. The accurate prediction of binding sites allowed by disordered regions of proteins is key to better understand the processes involved in protein interaction, which is of extreme importance for a variety of fields, that could help to eradicate diseases correlated to non functioning proteins or their altered availability. The overall goal of my project can be divided into two main sections. In the first part, a new dataset of proteins will be developed, with the intention to group all the proteins that contain binding regions, such as SLiMs or MoRFs. The objective of the second part of the project is to develop a new machine learning model for predicting protein regions prone to interactions starting from the protein sequence. The choice for the machine learning method is not straightforward, but in recent years there has been quite a significant improvement in the prediction of structural features from protein sequences using bidirectional recurrent neural networks (BRNN). The exploration and choice of the type of network appropriate for this task is one of the goals of this project, as it can serve as a template for other types of related works.