MutationTaster: automatic vs score

Question

[Farrel]

Why would I care if the classification into disease vs polymorphism was automatic or not? Wouldn't the probability of the model being correct all that I need?

I am looking at whole exome sequence data in a sample of affected subjects with a rare disease. The sequences were compared to GRCH38.p2 107 and variant files were derived and then run through annovar. I can see that I have two variables related to MutationTaster: MutationTaster_score and MutationTaster_pred.

the score is the probability of the model being correct from 0 to 1
the prediction classifies as four possible predictions:

1. "A" ("disease_causing_automatic")
2. "D" ("disease_causing")
3. "N" ("polymorphism")
4. "P" ("polymorphism_automatic")

I read http://www.mutationtaster.org/info/documentation.html but am none the wiser.

Answer 1

[Bioaln]

Normally, this distinction occurs when there exist entries, manually annotated by a human being, and entries, which are the result of an automated workflow. If this is the case, I suggest you place your bets on the manual thing. (example: SwissProt vs. UniProt - TrEMBL)

[Farrel]

Thanks for orienting me. Here is a hyperlink to illustrate Swiss-Prot vs. TrEMBL.

[Farrel]

I found a slideshow online that states exactly the opposite. Look at slide number 10.

Based on the answer by @Bioaln and based on the reference at Swiss-Prot it appears as if Hina Qaiser's slides are not correct. Do you agree or can you disavow me of my misconception.

[Bioaln]

There remains no problem, I interpret automatic as: well this protein is known to be in interaction etc., this is only semantics imho