search

phenopackets / phenopacket-format

26 stars 10 forks source link

Update requirements.md #27

Closed jmcmurry closed 8 years ago

jmcmurry commented 8 years ago

Clarifying comment: "Note this context must be explicit. For example, phenotypes that are mentioned in the "background" section of a journal article should not be included in the phenopacket unless they are also phenotypes observed in the patients/organism/gene under study. Otherwise, the algorithms used to analyze the phenopackets could infer nonsense associations."

jmcmurry commented 8 years ago

I hope that it is still clear that negation is OK? I just came across the scenario where background information wasn't a perfect match for the patient's phenotypes and thought it would be useful to clarify that we're interested in the asserted G2P relationships, not in just "tagging" journal articles.

"Ocular and genital abnormalities and cardiac defects have also been reported [in 2q31.1 microdeletion syndrome]" However genital abnormalities were not among the phenotypes reported for the particular patient in question.