phbradley / alphafold_finetune

Python code for fine-tuning AlphaFold to perform protein-peptide binding predictions
Apache License 2.0
137 stars 19 forks source link

Full datasets for training and testing #1

Open ilyes495 opened 2 years ago

ilyes495 commented 2 years ago

Hi, thank you so much for this great work! I know that the full dataset will be made available prior to the manuscript publication, is it possible to give us an estimate time of when it will be made available? if this will happen in months, can you then help us download the dataset on our own?

Thanks!

phbradley commented 2 years ago

Hi there, thanks for your interest! This should be ready, at least in preliminary form, in the next week or two. If longer, I will put an interim file up on dropbox and post a link here.

phbradley commented 2 years ago

Hi there, sorry for the delays. We are having a terrible time uploading the files to zenodo-- neither the web interface nor the command line/python interface seem to be able to handle the file sizes, so we are momentarily stuck. In the meantime, I uploaded a preliminary dataset with the fine-tuned parameters and the training and validation datasets to dropbox:

https://www.dropbox.com/s/k4gay3hwyq3k0rb/datasets_alphafold_finetune_v1_2022-08-02.tgz?dl=0

ilyes495 commented 2 years ago

Awsome! Thank you so much for your great efforts.

phbradley commented 2 years ago

Once you download the .tgz file, copy it into the alphafold_finetune/ folder and uncompress it, something like

tar -xzvf datasets_alphafold_finetune_v1_2022-08-02.tgz

that should create a new folder called datasets_alphafold_finetune/ and hopefully the other examples will work. Let me know if you run into trouble!

LivC93 commented 2 years ago

Any news on the full dataset? Since the weights are released, just the full test set would suffice.