Forum:Hot questions in sequencing data accessability

Question

[Tobias]

Currently. I am wondering what are the hottest open questions for working efficiently with large-scale sequencing data and making them accessible to either the public or some private collaborators.

Here I want to focus on solutions which (a) seem realizable (it would be great if a brief idea is presented) and (b) can be offered online for use of the community, such that everyone in a project can work with the data online.

Many thanks for your support in advance!

Answer 1

[Chris Miller]

The most pressing issues from my perspective are:

1. restrictive consents, which limit sharing
2. Lack of incentives to share cohorts and data. It's often better to hoard it, from the perspective of the institution and PI (if it was not collected/produced with NIH funding)

Neither of these are well-suited to informatic solutions :)

In general, there has been much progress towards moving data that is accessible to the cloud. Getting existing pipelines and tools ported to that infrastructure is a current source of friction (as is figuring out how to budget in usage payments to Amazon/Google, etc)

Answer 2

[Istvan Albert]

One major issue that I never recognized until taking on data re-analysis projects is how fragmented and untenable the current data "sharing" and distribution are. Some of the results will be distributed from the publisher's site via randomly named files called "Supplementary Information". These files could pretty much be anything, Excel files with multiple sheets,or tables embedded in a PDF file. Whereas the rest of data is deposited in SRA and does not even mention that more data can be found in the Supplementary information.