PMkbase is a web based bacterial phenotypic database based on the BiOLOG Phenotype MicroArray. The database reports phenotypes for various strain types and modifications based on a 4 step workflow of 1) Curation, 2) Quality Check, 3) Parameter extraction and 4) Growth calls. These phenotypes can be leveraged to A) Curate, gapfill and validate genome scale metabolic models, B) Look for interesting nutritional sources for generation of unique media conditions, C) Gaining phenotypic insights into the pangenome of a specie etc.
PMkbase is free for non-commerical use and is hosted and maintained at the Systems Biology Research Group at the University of California San Diego.
To submit any data you have, contact kkrishnan@ucsd.edu
High quality data should include samples with well annotated strain names/IDs, the specie they belong to, any genomic modifications done to it and the media conditions used (usually minimal media for metabolic plates). Atleast 2 replicates for a strain and plate combination are required to make a conclusive growth call
We use our in-house pipeline avaiable here, to make growth/no growth calls. First, we denoise the fluoroscence signals using a Savgol filter and generate the following kinetic parameters for every well in a plate + strain sample : 1) Maximum Respiration observed, 2) Maximum Rate of Respiration, 3) Time taken to get to maximum respirartion, 4) Area under the curve. A z-test is run on the Max Respiration observed in each well against all control wells to assign growth/no growth. A substrate/inhibtor is assigned as a trait to a strain if majority replicates display growth.