search

sina-mansour / UKB-connectomics

This repository will host scripts used to map structural and functional brain connectivity matrices for the UK biobank dataset.
https://www.biorxiv.org/content/10.1101/2023.03.10.532036v1
62 stars 7 forks source link

Missing rotated bvecs #33

Closed sina-mansour closed 2 years ago

sina-mansour commented 2 years ago

Nearly a fourth of the UKB subjects are missing rotated bvecs after eddy correction ("dMRI/dMRI/data.eddy_rotated_bvecs").

We're currently waiting for a response from UKB regarding the potential cause of this issue. But in the case that these aren't provided a solution (not to lose a fourth of the participant's data) is to use the raw bvecs files that are in fact available.

AndrewZalesky commented 2 years ago

Hi Sina,

we have not received a response from UKB about gradient rotation despite several follow up emails. It has been almost a month.

I think that we should proceed with using the uncorrected/raw bvecs to avoid further delay in my opinion. No idea when (or if) the UKB will respond.

Andrew

On 24/04/2022 4:43 pm, Sina Mansour L. wrote:

Nearly a fourth of the UKB subjects are missing rotated bvecs after eddy correction ("dMRI/dMRI/data.eddy_rotated_bvecs").

We're currently waiting for a response from UKB regarding the potential cause of this issue. But in the case that these aren't provided a solution (not to lose a fourth of the participant's data) is to use the raw bvecs files that are in fact available.

— Reply to this email directly, view it on GitHub https://github.com/sina-mansour/UKB-connectomics/issues/33, or unsubscribe https://github.com/notifications/unsubscribe-auth/ANXR7AVB2XZHIPXTMUZLSRTVGTURVANCNFSM5UFZU5MA. You are receiving this because you are subscribed to this thread.Message ID: @.***>

sina-mansour commented 2 years ago

Hi Andrew,

I agree, I'll change the script so that it only uses the eddy rotated bvecs if available.

I'm currently trying to resolve an issue about the expiry of our download key to access UK biobank data (I've contacted Ye to check whether a new key was issued).

Once that's resolved, I will test the new script and close this issue.

sina-mansour commented 2 years ago

Copying in the response we had from UKB admins here:

Thank you for your messages, and for your patience while we looked into this issue. The rotated bvecs from eddy were added to as additional eddy outputs into the diffusion packages as an update prior to 2020. These have not been used and should not be expected to make a noticeable difference from those without it. Evaluation of these files is expected for version v2 of the pipeline (in preparation). Further details may be available via the Neuroimaging mailing list – UKB-NEUROIMAGING@JISCMAIL.AC.UK which is monitored by the researchers who produced the neuroimaging pipeline.

Based on this response, I'll revert the code back to using the bvecs instead of the rotated ones for all subjects.