Karollus / SequenceModelBenchmark

MIT License
8 stars 2 forks source link

Request for tsv files #4

Open shhhhhh2001 opened 1 year ago

shhhhhh2001 commented 1 year ago

Hi! I'm a research assistant from the University of Michigan bioinfo lab. Since our lab also focuses on sequence-based predictive models, I'd like to replicate some analyses, specifically Enhancer Knockdown session performed in your paper, using the notebook you provided. However, I have encountered an issue with certain files such as 'avsec_fulltable_fixed-enformer-latest_results.tsv' missing in the Zenodo repository. I am writing to inquire whether these files may be available elsewhere, and if so, I would greatly appreciate it if you could provide me with the repository link. Alternatively, if these files are not accessible through any repository, would you mind providing me the complete set of files used in your study? Thank you.

Karollus commented 1 year ago

Hi,

Thank you for your interest! Unfortunately Zenodo has a 50GB file limit, so I was not able to place everything there.

I will send you an email with a google drive link where you can download the "avsec_fulltable_fixed-enformer-latest_results.tsv". This file is fortunately just 800mb.

Providing the complete set of files used in the study is a bit more difficult. There are overall >1TB of Enformer predictions, which goes beyond the limit of any free cloud service I know. If there are other specific files you need I can try to provide them, otherwise you can also run the Enformer predictions with the snakemake script, based on the input data files in the Zenodo.