Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Develop revert pre balance train #29

Merged
merged 34 commits into from
Jun 7, 2021
Merged

Conversation

FaizRizvi
Copy link
Member

Revert changes from Balanced Training to PCPC_Random Training.

Cazares, Tareian and others added 30 commits March 1, 2021 12:07
Plot AUPRC with thresholds. Added in a for loop to go through the input chroms for benchmarking. Will output a unique file per chromosome AUPRC.
Removed average signal from analysis and tested training and prediction.
This code should allow for either a BED file input for predictions or the names of the chromosomes you want to predict in.
#19 Whole genome prediction, ROI prediction, and chromosome predictions
There was an error for whole-genome prediction that caused it to crash after chromosome9 because it was not in the expected order of the header of the bigwig file. I fixed the chrom sizes file and sorted it by number correctly. I also removed the sorted() function that ordered the chromosomes incorrectly.
The call peaks function ended up producing the wrong peaks for our models because it as subtracting a 1 from the start position. This has been fixed.
When we import the data for prediction we are not replacing None values with 0. This is different than the approach that we used with bulk data analysis. The reason we want to do this is because we are import None values and I am not sure how that changes our predictions.
Added an optional argument to the parser for setting the min and max values you want to normalize the scores to.
Fix the bugs with string - int type error
@tacazares tacazares merged commit 8aa9df9 into main Jun 7, 2021
@tacazares tacazares deleted the develop_revert_pre_balance_train branch June 7, 2021 19:43
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants