diff --git a/README.md b/README.md index 3143312..7286582 100644 --- a/README.md +++ b/README.md @@ -49,7 +49,7 @@ wget http://purl.obolibrary.org/obo/hp.owl So, to run the program minimally, you can type: ```bash -klarigi --data data.txt --ontology hp.owl --group OMIM:604271 +./klarigi --data data.txt --ontology hp.owl --group OMIM:604271 ``` The *--data* argument is the path to the file containing entities/groupings (as above), *--ontology* is the path to the ontology @@ -78,3 +78,7 @@ Danger Warning: if there are unmapped values in your information content file, t Todo: Currently the IC generation method uses the Zhou method, and it is not configurable. SML supports many other methods, however, so I plan to add additional configurability. +#### Group + +We can use the *--group* argument to tell it which grouping to provide an explanation for. If you don't pass this argument, or pass *--group \**, then it will give you explanations for all of the groups described in your data file. +