Oren Tzfadia - ITG Antwerpen: Functional Bioinformatics: identifying new candidate genes in (metabolic) pathways
Daniele Parisi - KU Leuven: Netflix and chill
About the Lunch Talks
The Biomina Lunch Talks are an initiative of a number of young researchers in the biomina network and is sponsored by the Flemish Government. We aim to stimulate the interaction between researchers from different disciplines who encounter bioinformatics and computational biology, and consequently we focus on a broad and multidisciplinary public. With this informal medium we would like to provide a platform where knowledge and experience can be presented and exchanged, across partners from both academia and industry. In this manner we have had the pleasure to welcome speakers from various institutes such as the University of Antwerp, the Institute of Tropical Medicine, Janssen Pharmaceutica, the Antwerp University Hospital and Open Analytics. Last, but not least, these sessions can provide a great opportunity for young researchers to acquaint themselves with new ideas and methods in the field of bioinformatics and medical informatics.
Oren Tzfadia, ITG Antwerpen
Title: Functional Bioinformatics: identifying new candidate genes in (metabolic) pathways
Abstract: Closing gaps in our current knowledge about biological pathways is a fundamental challenge. The development of novel computational methods along with high throughput experimental data carries the promise to help in the challenge. I will present a set of novel algorithms for revealing unknown genes in biological pathways. These methods utilize known genes from the target pathway, a collection of expression profiles, and interaction and metabolic networks, using co-expression and machine learning techniques, to output ranking of candidate genes predicted to belong to the target pathway. The methods are generic and can be applied to any other species with appropriate data.
Daniele Parisi, KU Leuven
Title: Netflix and chill
Abstract: Identifying drug-target interactions is a crucial step in drug repositioning, the process of suggesting new indications for known drugs. There are about 9000 FDA-approved and experimental small molecule drugs and more than 500.000 protein records available. Performing in vitro experiments would be too expensive and time-consuming to check all the putative drug-target couples, therefore computational techniques might help to predict compound biological activity (IC50) and suggest new putative medical indications for existing drugs. Similarly for what happened for the "Netflix Challenge" (the attempt to predict the couples user-movie from Netflix based on the scarcely filled matrix of movie rating), given a drug-protein interaction matrix with a fill rate of ~1% it's possible to predict new values from given data using Scarce Matrix factorization coupled with supervised learning. Moreover, in the same way a dating app matches people based on their personal characteristics; Bayesian Matrix Factorization can integrate structural information of drugs, proteins and their binding to better predict the affinity of their interaction (IC50) and suggest new drug-target targets, with a big impact on the drug discovery process. Different kind of side information can be used to help the prediction process, such as chemical structures of the drugs, 3D structures of the protein targets or phenotypic effect of drug-target interactions. In this work we analyse the contribution brought by structural information in the prediction process, taking into account the difficulties related to the usage of those kind of data.