Validation Data Missing Epitope Label

[sidhomj]

The current file with the validation cohort does not have the epitope label. It instead has EPI for all clones.

[phbradley]

Hi there-- thanks for pointing this out. As I recall, for this dataset not all of the TCRs are specific for one of the four epitopes; these cells were stained with four tetramers and index sorted, so we have MFI values for all four (the *_MFI fields toward the end). Given a threshold on those MFI values, one can define a set of positive binders. For the majority of the original dataset we did not have this multi-tetramer MFI information, which explains why this file is structured a little differently. Let me know if that's not clear. Take care, Phil.

[sidhomj]

Thanks! I figured it out eventually. Should have closed the issue. Also, does tcr-dist take amino acid sequences directly? Or only nuc?

JW

Sent from my iPhone

On Feb 26, 2019, at 10:01 AM, Philip Bradley notifications@github.com<mailto:notifications@github.com> wrote:

Hi there-- thanks for pointing this out. As I recall, for this dataset not all of the TCRs are specific for one of the four epitopes; these cells were stained with four tetramers and index sorted, so we have MFI values for all four (the *_MFI fields toward the end). Given a threshold on those MFI values, one can define a set of positive binders. For the majority of the original dataset we did not have this multi-tetramer MFI information, which explains why this file is structured a little differently. Let me know if that's not clear. Take care, Phil.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/phbradley/tcr-dist/issues/29#issuecomment-467471371, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AarMzJxPvfBAK_FeD5oKwIu23V3vTKBqks5vRUwQgaJpZM4bRdUp.