Scientific Seminar with Invited Speaker
Daniela Kalafatović, University of Rijeka
When |
26 Nov, 2021
from
11:00 am to 12:00 pm |
---|---|
Where | ITQB NOVA Auditorium |
Contact Name | Ana Pina |
Add event to your calendar | iCal |
Title: Design of short peptides and peptide assemblies aided by machine learning and genetic algorithms
Speaker: Daniela Kalafatović
Affiliation: Department of Biotechnology, University of Rijeka, Croatia
Abstract:
The discovery of new peptides with desired properties based on their sequence is challenging, as they are part of a very large search space and the principles responsible for the desired properties are not yet fully understood. To avoid expensive and time-consuming guesswork and experimental failure, our strategy is to apply soft computing techniques to address these issues. Search based algorithms allow for a faster exploration of peptide permutation space which grows exponentially with peptide length and whose amount and dimensionality is too overwhelming to rationally comprehend. Machine learning can find patterns or regularities in data, build mathematical models based on the theory of statistics and make up for the lack of knowledge. To date, soft computing has been applied to a variety of chemical problems to maximize the chance of successful and rapid solving of complex issues.
Our team has already reported on a multi-objective evolutionary approach for the exploration of mass and sequence diversity-oriented random peptide libraries and on the bottom-up approach for their design, which combines physico-chemical properties obtained experimentally and theoretically to cover larger parts of the peptide chemical space. Our current activities involve the application of soft computing techniques to find peptides with catalytic activity, to predict their predisposition towards self-assembly and to estimate the antiviral activity of selected candidates (figure 1). The aim behind these objectives is to rationally design supramolecular peptide-based viral mimetics of SARS-CoV-2. Peptide-based materials can resemble the morphology of the viral capsid while offering the possibility to introduce tunability, modularity and responsiveness to stimuli.
Short Bio:
Daniela Kalafatović obtained her PhD in Pure and applied Chemistry at the University of Strathclyde, Glasgow, in 2014.Between 2016 and 2018 she was a Postdoctoral fellow under the Marie-Curie Actions, at Institute for Research in Biomedicine, Spain. She is currently Head of Medicinal Chemistry Division and Assistant Professor at the University of Rijeka and is working on a project of Design of short catalytic peptides and peptide assemblies.