Clarification Needed on dbNSFP Plugin Annotation for Variants with Multiple Values

[sibylZHAI]

Dear Developer,

I hope this message finds you well. I am currently using the VEP Plugin dbNSFP to annotate my VCF file. However, I have encountered a peculiar situation where annotations for some variants display multiple values separated by an "&" symbol. These values are sometimes equal and sometimes not, and they also include the "." character.

For your reference, I have attached a table illustrating the issue. Could you please explain why this discrepancy occurs and how it can be resolved?

I would greatly appreciate any guidance or insights you could provide on this matter.

Thank you for your time and assistance.

Best regards,

Zhai

AlphaMissense_pred | CADD_phred | Polyphen2_HDIV_pred -- | -- | -- B&B | 2.209 | .&B B | 14.21 | B A | 24.7 | D B | 6.599 | B B | 9.609 | null P&.&A&P&P&A&P | 27.3 | D&.&.&D&.&.&. B&B&B | 23 | .&B&. B | 13.08 | P B | 0.926 | B B | 13.64 | D B&B | 0.017 | .&. B&B | 4.449 | .&. B | 12.7 | B B | 10.58 | B

[dglemos]

Hi @sibylZHAI, Most of the dbNSFP scores are transcript-specific.

Example For variant: 1 999169 A/C

dbNSFP 4.7c

Ensembl_transcriptid    ENST00000428771;ENST00000304952;ENST00000484667
AlphaMissense_score    0.1133;0.113;0.1316

(; separates different values)

For this example there are 3 Alpha Missense scores, one for each transcript: ENST00000428771 -> 0.1133 ENST00000304952 -> 0.113 ENST00000484667 -> 0.1316

In your example, B&B is a score for each transcript. If you check the column Ensembl_transcriptid you should have two transcript IDs.

Here you can find more information about dbNSFP:

• dbNSFP website: https://sites.google.com/site/jpopgen/dbNSFP
• dbNSFP publication: dbNSFP v4: a comprehensive database of transcript-specific functional predictions and annotations for human nonsynonymous and splice-site SNVs

Let me know if you have more questions.

Best wishes, Diana

[sibylZHAI]

Hi Diana,

Thank you for the detailed explanation and resources. This helps a lot!

Wishing you the best, Zhai