Chloroplast 2010: A Database for Large-Scale Phenotypic Screening of Arabidopsis Mutants

Large-scale phenotypic screening presents challenges and opportunities not encountered in typical forward- or reverse-genetics projects. We describe a modular database and Laboratory Information Management System (LIMS) that was implemented in support of the Chloroplast 2010 Project, an Arabidopsis reverse genetics phenotypic screen of >5,000 mutants (http://bioinfo.bch.msu.edu/2010_LIMS; www.plastid.msu.edu). The software and laboratory work environment were designed to minimize operator error and detect systematic process errors. The database uses Ruby on Rails and Flash technologies to present complex quantitative and qualitative data and pedigree information in a flexible user interface. Examples are presented where the database was used to find opportunities for process changes that improved data quality. We also describe the use of the data analysis tools to discover mutants defective in enzymes of Leu catabolism (heteromeric mitochondrial 3-methylcrotonyl CoA carboxylase, At1g03090 and At4g34030; putative hydroxymethylglutaryl (HMG)-CoA lyase, At2g26800) based upon a syndrome of pleiotropic seed amino acid phenotypes that resembles previously described ivd1 (isovaleryl dehydrogenase, At3g45300) mutants. In vitro assay results support the computational annotation of At2g26800 as HMG-CoA lyase.