The MSR is a statistical parameter that characterizes the reproducibility of potency estimates from in vitro concentration-response assays. Biological activity expressed as potency of compounds is an important parameter in screening and drug discovery. This app allows users to visualise the Bland-Altman plot of biochemical and cellular data stored in the DM oracle database.
There are three Oracle Functions that are of interest: 1) CALC_MSR2 2) MOST_RECENT_FT_NBRS2 3) GEN_GEOMEAN_CURVE_TBL. CALC_MSR2 relies on MOST_RECENT_FT_NBRS2. The full code can be viewed by observing the Functions tab within SQL Developer. Function 3) generates the Curve QC tool output that is triggered as a pop-up within DM. Note the same functions without the ‘2’ suffix are deprecated functions that aren’t used and was developed during testing phase. Function 3 now takes significantly longer to process the SQL query since it will calculate every unique combination of the following fields: CRO, assay type, variant, (cell incubation hr / target), (atp conc / pct serum), (cofactors / cell line). However a dictionary is generated to keep track of each of these unique combinations, and would reduce the time to render. Astrix Technologies consultatnts have developed an additional layer to keep track of these unique combinations in an Oracle table and update overnight. This helps with global caching and should reduce the time to render significantly if the combination has been encountered in the past and stored in the table, MSR_CALC.
The MSR plot piggy-backs on the geomean-ic50-flagger app that was written in ReactJS+python. Since ~30% of the code can be re-used it was decided to append a separate module to the geomean app rather than create a new one from scratch and duplicate code. Function 3 from above passes in query string parameters to provide to the JavaScript application that then calls the correct API to the python backend app.