A neural network model for Lead Optimization of MMP12 inhibitors

Tewodros M. Dagnew


20 April 2020, 10 a.m

Online Seminar 


Lead Optimization is a complex process, whereby a large number of interacting entities give rise
to molecular structures whose properties should be optimized in order to be considered for
drug development. We will study molecular systems that are characterized by high
dimensionality and dynamically interacting networks. Currently, the research involves screening
and identification of molecule subsets from a large molecule library with desirable response-
properties. Lead Optimization is a multi-objective optimization problem. The classical approach
of in-vitro laboratory analysis is time taking and very expensive. To address this problem, we
propose an in-silico approach: Lead Optimization based on neural network (NN) model in order
to help the chemist in the process by requiring a small set of real laboratory tests. In this paper,
we propose and estimate a predictive model to derive a simultaneous optimal multi-response
property following a multi and single objective optimization procedure. We adopt two different
architectures in this simulation study and we compare our procedure with other state-of-the-
art method showing the better performance of our approach.


Bio sketch

Tewodros M. Dagnew joined ECLT in 2019 as a postdoc researcher under the supervision of Prof. Irene
Poli to work on the topics of predictive models for high dimensional and complex data specifically on
projects revolving around personalized and precise medicine and lead molecule optimization in the
context of drug discovery.
He received his Ph.D degree (2019) in Computer Science (CS) from University of Milan, with a thesis
titled ‘Machine-Learning based analysis and computer aided classification of neuropsychiatric-disorders
using neuro-imaging’ under the supervision of Prof. Sassi and Prof. Brambilla (MD). Before that, he has
been a researcher in University of Verona working on Computer Vision area under the supervision of
Prof. Castellani. He received his MSc. degree (2013) in CS from Ca’Foscari University of Venice under
the supervision of Prof. Pelillo and his BSc. degree (2009) in CS from Jimma University under the
supervision of Geletaw Sahle.