Invited Talk in Causal Discovery from Mass Cytometry Data

Our group is presenting our latest discoveries and newly developed Causal Discovery algorithms in the workshop “Case Studies of Causal Discovery with Model Search”. The presentation will be recorded and be made publicly available. Some of this work stems from a close collaboration with our Karolinska Institutet STATegra partners in the Computational Medicine Unit. The workshop includes an impressive panel of invited speakers, held in one of the “birthplaces” of Causal Discovery. We hope to see and personally meet many Causality enthusiasts in the workshop.

First Year STATegra Project Meeting

Our group participated in the first year’s STATegra project meeting, held in the Imperial College London. The meeting was a great success. Novel data will be available soon, while new methods and tools are continuously being developed by the consortium. New upcoming events will be held this March. STATegra has a new, updated web-site containing – among others – information about upcoming events (stategra.eu).

New papers in Cell Metabolism and Nucleic Acids Research

This year, our group’s research has transcended the computer science and bioinformatics publishing forums and, through collaborations with prestigious biological groups, has participated in publications in high-impact journals of the biology community. The Garini’s lab research has shown that that DNA damage triggers a chronic auto-inflammatory response leading to fat depletion published in Cell Metabolism. The Strouboulis lab studied the role of GATA-1 in mouse fetal liver erythropoiesis published in Nucleid Acids Research.

Research Staff Positions and Graduate Fellowships

BIL has recently secured funding in competitive and exciting EU and Greek National research grants (BIOSYS by GSRT, eModia by GSRT, STATegra EU FP7, as well as participation in other grants). BIL is looking to expand by hiring ambitious, energetic, intelligent, and hard-working individuals to lead the development of novel machine learning methods, particularly for bioinformatics and computational biology problems, as well as innovative application of bioinformatics to biological data. There are several positions available; some of them have formal procedures for hiring depending on the source of the funding. Please, check the procedure carefully before applying.

For further information follow this link

New institutional project on Systems Biology

Our group is participating in the new institutional grant on Systems Biology, named BIOSYS, coordinated by the Institute of Molecular Biology and Biotechnology of FORTH. The grant will provide more funding for developing causal discovery algorithms, as well as model selection algorithms for machine learning models. The algorithms will be packaged in tools for non-experts and be applicable to common omics biological data.

Introduction to Causality Tutorial at the High-Throughput Omics and…

Members of our group gave a tutorial on Causal Discovery, titled “An Introduction to Causal Discovery, a Bayesian Network Approach” in the recent the High-Throughput Omics and Data Integration Workshop organized by the STATegra projects and the COST Action SeqAHead. Apart from the introduction to the theory and algorithms of Causal Discovery, the tutorial covered case studies in biology where causal analysis algorithms have led to new discoveries.

Best Poster Award at HSCBB 2012

Sofia Triantafillou received one of the three Best Poster Awards in the recent 7th Conference of the Hellenic Society for Computational Biology and Bioinformatics (HSCBB12) for her poster titled “Predicting associations from multiple “omics” data sets”. The judging committee of the conference consisted of all the invited speakers of the conference.

New positions available – Job Announcements

New positions are available at FORTH within our group. We have secured funding in competitive and exciting EU research grants and are looking to expand by hiring ambitious, energetic, intelligent, and hard-working individuals to lead the development of novel causal discovery and machine learning methods, particularly for bioinformatics and computational biology problems, as well as innovative application of bioinformatics to biological data. Post-doc position also advertized at www.KDnuggets.com – Analytics and Data Mining Resources

Link to the detailed announcement

New versions of our toolboxes

Mens X Machina Probabilistic Graphical Model Toolbox 0.9.3 and Mens X Machina Common Toolbox (previously named Commons Toolbox) 0.9.3 are available for download. The toolboxes were completely rewritten and are mostly object-oriented now. Support for the MMHC algorithm and the junction-tree algorithm for structure learning and inference, respectively, was added. The toolboxes are not compatible with their previous versions; please see the release notes for a list of the most important changes if you wish to migrate to the new versions.

Mens X Machina Bayesian Network Samples (previously named Probabilistic Graphical Model Samples) 2.0 is also available for download. Samples from the Andes, Diabetes, Link, Mildew, Munin, Pathfinder, Pigs, Powerplant and Water Bayesian networks were added.

Incorporating Prior Causal Knowledge when learning Causal Models in…

Knowledge in the form “A causes B” and “A does not cause B” is often available. For example, in an experiment where temperature (A) is controlled, then for any quantity B that changes we can infer A causes B. Our new publication in ICML shows how to incorporate such knowledge when inducing a causal models from other datasets.