search

monarch-initiative / pheval

A framework for empirical evaluation of phenotype matching and prioritisation
https://monarch-initiative.github.io/pheval/
Apache License 2.0
12 stars 1 forks source link

Develop a data comparison between two (phenotypic) semantic similarity profiles #6

Closed souzadevinicius closed 1 year ago

souzadevinicius commented 2 years ago

Motivation

In order to study the effects of phenotypic semantic similarity profiles we need to first characterize the difference. See also #4

Implementation

Develop a python function as part of a Jupyter notebook that takes as input two pandas dataframes with semantic similarity profiles and analyses how they differ from each other. The metrics to describe this difference need to be developed as a first pass we want to understand what is the average change of all phenotypic similarity values across pairs of ontology classes. For example, consider SSP1, with N:1,N:2,0.9 and SSP2 with N:1,N:2,0.45 the percentage change is -100% (google it!)

Testing

Use the original HPHP mapping table from exomiser and compare it with the scrambled results from #4.

matentzn commented 2 years ago

Track: https://github.com/Knowledge-Graph-Hub/semsim/pull/12

souzadevinicius commented 1 year ago

Perhaps this issue can be closed after PR #48 merge