Privacy issues for proteomics data

Privacy issues for proteomics data

Overview

[Slide by Henning Hermjakob to introduce the problem.] (https://github.com/CompMS/Overview/blob/master/privacyInProteomics-hhe.pdf)

Essentially, proteomics data is currently not considered as having privacy issue (goes as far as FASTA databases derived from personal sequencing, which are not considered an issue today)!

But two recent papers have highlighted that privacy issues can occur for proteomics data: Li et al, 2016 and Parker et al, 2016.

Also, current MS proteomics data is not interpreted in full. Most notably, modification patterns or single aminoacid polymorphisms (SAAPs) are typically left unidentified. If we however, tomorrow have a new tool that can identify these, existing data sets might become retroactively privacy sensitive.

The best possible approach right now, is to be aware of possible future risks, and start asking for broad consent, modelled on how this is done for genomics. This will prepare the field for future emerging risks.

HUPO should start spreading awareness about this, and work with clinicians, ethicists, and expertise from genomics, to come up with consent forms.

HUPO-PSI could have a new Working Group around this.

There is an ethics committee at the HUPO Council level, but it is more or less on hold. We can reactivate this.