D-Score Significance

Screening assay results are calculated using a Pipeline Pilot implementation of a procedure developed by Paul Clemons and his research group [PMID:24464433]. This specific implementation was developed by Paul Clemons, Joshua Bittker, Karen Emmith, and Joshua Sacher. Screening assay results report the D-score, which is a dimensionless z-score-like measure of significance, based on the effect size, and normalized by appropriately weighted variance contributions across replicates.

BARD Ontology

BioAssay Research Database [PMID:25477388] Ontology – Developed as part of the public repository of the NIH Molecular Libraries Program, the ontology is a curated, consistent vocabulary that includes a standard representation and common language for organizing bioassays and their results. This curation enables systematic, cross-assay analysis of datasets regardless of the identity of the contributing center or individual data submitter. BARD uses a controlled vocabulary specifically developed to describe assay protocols.

How to Use

The Repurposing Assay Results Viewer will display D-Score results for all Repurposing Hub compounds as presented in a heat map grid format using the Morpheus visualization tool developed by Joshua Gould at the Broad Institute.

1. Assays
   1. Selecting Assays

      Assays are listed by name along the top of the visualization. Clicking an assay will highlight the column of data underneath for the entire height of the grid. To select multiple assays you can hold Shift or Control while left-clicking an assay name.

   2. Sorting Assays

      The rows directly underneath the assay names represent assay properties, the values of which can be used to sort and group assays. Clicking on an assay property will apply a sort according to the values in the selected row. Holding Shift or Control while left-clicking an assay property name will select additional assay properties by which to sort.

   3. Filtering Assays

      Selecting properties from the drop-down menu(s) in the Assay Filters panel will filter out any assays that do not match the chosen parameters. There is no limit on the number of filter values that can be selected.

2. Compounds
   1. Sorting Compounds

      The two columns to the far right of the grid show the Compound Name and the Broad Compound ID for each compound. Clicking on the header of one of these columns will sort the compound list alphabetically.

   2. Searching for Compounds

      Typing a search string into the text box at the top of the grid next to the “Columns” button and pressing enter will yield a set of matching results for your entry. The up and down arrows to the right of the search box will traverse the grid for matches. Pressing the third button that looks like an arrow with a right angle pointing up will move all matches to the top.

   3. Filtering Compounds

      Compound filters are grouped vertically along the left side of the grid. Selecting a compound property will filter out any compounds that do not match the selected properties. There is no limit to the number of compound filter properties that can be selected.

3. Result Data
   1. Viewing Results

      Moving the mouse cursor over a cell in the grid will show the significance, concentration, compound name, and assay for that data point. Clicking on it will populate a panel to the right side of the window that contains more detailed information. This detailed information includes the compound properties and a hyperlink to the Clue.io repurposing app details page for that compound.

   2. Exporting Results

      Moving the mouse cursor over a cell in the grid will show the significance, concentration, compound name, and assay for that data point. Clicking on it will populate a panel to the right side of the window that contains more detailed information. This detailed information includes the compound properties and a hyperlink to the Clue.io repurposing app details page for that compound.