Organizers of the 12th annual Conference on Critical Assessment of Massive Data Analysis have launched two competitive challenges that aim to assess the effectiveness of computational methods used to extract useful information from massive heterogeneous datasets.
The first challenge asks participants to use a dataset from the Japanese toxicogenomics project to predict drug compatibility. The organizers have provided rat liver data from the project, which they expect participants to analyze to explore questions such as whether in vitro assays can replace animal models in safety studies or whether it's possible to predict drug-induced liver injury in humans from animal-based toxicology studies.
The data package for this challenge includes array data, drug information including annotation data and labels, pathology data, and more.
The second challenge calls for participants to analyze human genome data from the Korea Personal Genome Project. Its goal is to explore methods of analyzing and utilizing next-generation sequencing data and electronic medical records to improve patient care.
For this challenge, participants have been provided with 38 human genomes sequenced to 30x to 40x coverage on an Illumina HiSeq 2000 platform, as well as limited medical record data on the study participants.
The submitted methods are expected to address questions such as how to call single nucleotide variants and distinguish them from sequencing errors; how to reliably call structural variants; how to identify identical DNA segments that indicate shared ancestry; and more.
Interested participants are expected to submit abstracts for an oral presentation of their methods by May 20. They are also expected to submit poster abstracts by May 25.
Additional information about the challenge and the CAMDA conference, which is a satellite meeting of the annual conference on Intelligent Systems for Molecular Biology, is available here.
This year's ISMB conference will be held from July 19 – 23 in Berlin, Germany.