Click here to visit Github repository with example input files and detailed instructions!
Input Requirements
📁 MAVE CSV Files
- Format: CSV files with variant effect scores. Each csv should be two columns only; each MAVE should have its own separate csv file.
-
Required Column:
hgvs_pro- variant identifier, ideally in p.REFposALT format, 3 letter or 1 letter amino acid code. NOTE: THE FORMAT USED HERE MUST MATCH IN ALL INPUT CSV FILES - Score Column: Column containing the MAVE score. Custom names with standard characters are fine. For example: 'PTENvamp_score' or 'PTENcellFitness_score
- Multiple Files: Upload 2 or more MAVE datasets for the same gene
📋 Truth Set CSV File
- Format: CSV file with known variant classifications
- Required Columns:
-
hgvs_pro- variant identifier (must match MAVE files) -
binary_clinvar_class- classification with values 'P' (pathogenic) or 'B' (benign) -
SYNorPTC- classification with values 'SYN' (synonymous) or 'PTC' (premature termination codon) ... please check that 'SYN' variants are also labeled 'B' and that 'PTC' variants are also labeled 'P' in the binary_clinvar_class column
⚙️ Workflow
- Upload your MAVE CSV files
- Upload your truth set CSV file
- Click 'Perform Analysis'
- View results and download outputs