You are here

Significant progress in data analysis that increases discoveries in how genes contribute to phenotypes

We are pleased to announce the development of a new machine learning approach to data analysis that is robust, more sensitivity and increases our understanding of gene-phenotype relationships.  The method, recently published in PLoS One (, is adaptive to the complex data collected in high-throughput phenotyping pipelines.  Applying this analysis method further optimizes the phenotyping work flow by using knockout mice in small independent batches and comparing this to a baseline control dataset.  These changes have significant economical and ethical benefits as it reduces the use of animals, increases throughput, and decreases cost while improving the quality and depth of knowledge gained.  Further work will refine the method to encompass all data types from the phenotyping process, validate the output and develop interpretation tools for biologists.  These tools combined with IMPC data will create an unparalleled resource for gene function that will fuel future discoveries in mammalian and biomedical sciences.

PDF icon journal.pone_.0052410.pdf