Open pnrobinson opened 2 years ago
@iimpulse - in case you want some more inspiration?
I wonder if you would find it useful to allow sorting the phenotypic features in the output, e.g. to put the more similar features together. In principle, something like hierarchical sorting here.
The 'sort' order is currently the input order, but I like the idea of sorting like this. Alternatively ordering by high-level system under Phenotypic abnormality first, then by descending IC.
In recent work, we are getting on the order of 50-100 HPO terms per proband (from EHR text mining). This makes the current output of Exomiser unweildly. Even with smaller numbers of terms, the output could be improved. Here are some thoughts