PaccanaroLab

09/12/2021

We are excited to share our latest article: " for newly sequenced organisms", published today in Nature Machine Intelligence: https://rdcu.be/cCVsT

In this paper we introduce S2F (sequence to function). A machine learning method that predicts protein function for organisms that have only sequence information.

The twist? We go beyond homology to do it!

By transferring protein-protein relationships from well studied organisms, we put together a functionally coherent network of proteins.

We use InterPro and HMMER predictions as a starting point. And these predictions are propagated and amplified on the network.

In development for many years, the project is open source https://github.com/paccanarolab/s2f

(PRs are welcome!)

An interactive explorer of our extensive testing is available in the project website:
https://paccanarolab.org/s2f/

Predicting the function of proteins in newly sequenced organisms is a challenging problem. Mateo Torres et al. present here a method to transfer the functional relations from known organisms and improve the prediction using network diffusion.

23/11/2021

We are excited to share our on and approaches for for ! https://www.cell.com/patterns/fulltext/S2666-3899(21)00263-4 published in Cell Press

We present two complementary machine learning approaches for drug repositioning against COVID-19 that target SARS-CoV-2 and its cellular processes in the host, respectively. Our matrix decomposition approach exploits drug developmental information to predict broad-spectrum antivirals; our graph kern...

17/09/2020

Our recent paper "Predicting the frequencies of drug side effects" featured in TechXplore! 🤓😍🙌 Read our paper here: https://rdcu.be/b64BK

A new algorithm has been developed by academics at Royal Holloway, University of London, to predict the side effects of drugs before they hit the market by using the same principle by which movies are recommended to users.

11/09/2020

We are excited to share our on https://www.nature.com/articles/s41467-020-18305-y published in SpringerNature Nature Communications

Currently, the frequencies of drug side effects are determined in randomised controlled clinical trials. Here the authors develop an interpretable machine learning approach to predict the frequencies of unknown side effects for drugs with a small number of determined side effect frequencies.

07/07/2020

Everyone is invited to attend our webinar on AI-based Drug Repositioning for COVID-19 this Thursday!. We will present what we’ve been working on in the last couple of months! Please join us: https://bit.ly/2O2kMi6 (registration is required)

Quais os resultados preliminares nas áreas de medicina de rede e aprendizado de máquina no desenvolvimento de novos algoritmos para a previsão de medicamentos reposicionáveis para a Covid-19? O debate, liderado pelo pesquisador Alberto Paccanaro, reunirá os alunos de Pós-Doutorado da FGV EMAp para discutir o trabalho em andamento. O webinar será ministrado em inglês. Participe: https://bit.ly/2O2kMi6

22/07/2019

Our new paper is out! Congratulations to our lab member Dr Juan Caceres for achieving the milestone of obtaining his doctoral degree!

Author summary The elucidation of the genetic causes of diseases is central to understanding the mechanisms of action of a pathology and the development of treatments. Disease gene prediction methods streamline the discovery of the molecular basis for a disease by prioritizing genes for experimental...

06/09/2017

Prof. Alberto Paccanaro and his graduate student Ruben Jimenez were recently at the XLII Latin American Conference on Informatics (CLEI) & 46th Argentine Conference in Informatics (JAIIO) that was held in Cordoba, Argentina. PaccanaroLab research group presented three papers:
(i) Drug cocktail selection for the treatment of Chagas Disease: a multi-objective approach;
(ii) An exploratory analysis of drug target locality;
(iii) Mining the biomedical literature to predict shared drug targets in DrugBank.
Thanks for the invitation! We have a great time sharing our research!

02/06/2017

Today was the Computer Science Annual Research Colloquium at Royal Holloway, University of London. Juan Caceres, Mateo Torres and Diego Galeano presented their research in short talks. Namely, "Predicting genes for molecularly uncharacterised diseases", "Gene pushing" and "Signatures of drug side-effects in human phenotype", respectively. Congratulations to the guys for their successful viva, and for winning the prize of best presentation during the colloquium.

01/03/2017

We invite you to join us today at 4pm for the Computational Biology and Bioinformatics Seminar Series. Dr. Alberto Paccanaro will give a talk about Network theory applications to Biology problems at Yale University.

Title: “Answering questions in biology and medicine by making inferences on networks”
Date: Wednesday, March 1, 2017
Time: 4:00 seminar 5:00 refreshments
Place: BML Auditorium, 310 Cedar Street
Hosted by: Mark Gerstein, PhD

22/02/2017

Professor Alberto Paccanaro giving a talk at the Yale Institute for Network Science, YINS Seminar Series.

“Inference and structure discovery in protein interaction networks”

Abstract: An important idea that has emerged recently is that a cell can be viewed as a complex network of interrelating proteins, nucleic acids and other bio-molecules. At the same time, data generated by large-scale experiments often have a natural representation as networks such as protein-protein interaction networks, genetic interaction networks, co-expression networks. From a computational point of view, a central objective for systems biology is therefore the development of methods for making inferences and discovering structure in biological networks possibly using data which are also in the form of networks.

In this talk, I’ll present novel computational methods for solving biological problems which can all be phrased in terms of inference and structure discovery in large scale networks. These methods are based and extend recent developments in the areas of machine learning (particularly semi-supervised learning), graph theory and network science. I will show how these computational techniques can provide effective solutions for: de-noising large scale protein-protein interaction experiments; detecting protein complexes from protein-protein interaction data. Finally, I’ll describe how these ideas could be applied to problems in the area of Network Medicine, such as disease gene prediction and drug repositioning.

Bio: Alberto Paccanaro is Professor in Computational Biology at Royal Holloway, University of London. He completed his undergraduate studies in Computer Science at the University of Milan in 1990 and received his PhD from the University of Toronto in 2002, specializing in machine learning under the supervision of Geoffrey Hinton. From 2002 to 2006 he was a postdoc at Queen Mary University of London and Yale University. His research interests are in applying and developing machine learning and pattern recognition techniques for solving problems in molecular biology. His recent work has focused on the development of methods for analysis and inference in large scale biological networks.

Homepage: www.cs.rhul.ac.uk/~alberto

Lab page: www.paccanarolab.org

13/02/2017

PaccanaroLab team doing Research Visiting at Yale University in GersteinLab! It's been so far an amazing experience! Months to come of productive work and cheerful time with friends! Thanks to Cristina SD Cannon for hosting us ;) ! Diego Galeano Mateo Torres Juan Caceres Ye Cheng