Home

Welcome! copairs is a Python package that implements mAP framework for evaluating strength and similarity of high-dimensional profiling datasets.

Mean Average Precision (mAP) can be used to indicate the degree to which profiles from one group exhibit greater intra-group similarity compared to their similarity with the profiles from a second group. Thus, it can be used to compare any two groups of high-dimensional profiles, such as replicates of the same perturbation against controls, perturbations annotated with related phenotypic activities against other perturbations, one experimental batch of samples against another, etc.

copairs can be applied to any profiling datasets containing metadata and the corresponding feature values. Calculations are performed in 3 steps:

• defining groups of profiles for comparison using user-defined rules based on metadata
• assessing intra- and inter-group similarity by calculating the mAP score
• using permutation testing to determine the statistical significance of the mAP score

See more information on how to define parameters here.