Can't seem to download the dataset

[alexberlaga]

Hi! I tried to clone the dataset from the ModelScope repo as the README suggests, but it looks like there's no data in there. I just get another README file which is under 1KB. Am I missing something?

[atong01]

I also see no data

[sankethvedula]

I see that the ModelScope repository has some sample data (1ab1_A) -- the full dataset seems to still be missing. Am I missing something? Could you please also let me know where I can find a list of PDBs for which the simulations were run?

[h0ngxuanli]

Hi there! I noticed there are currently around 500 data in the repository. I was wondering if you could kindly let me know the expected timeline for completing the full dataset upload? This would help me better plan my work. Thank you in advance!

[zqcai19]

@h0ngxuanli Hi, thank you for your interest in our dataset.

Before initiating the data upload, we carefully evaluated multiple cloud storage solutions, considering both capacity and bandwidth limitations, and ultimately chose the current platform for its balance of performance and accessibility.

Due to the immense size of the dataset, we are currently able to upload data files for an average of six proteins per day. But we are doing our best to continuously upload the data.

We sincerely appreciate your patience and understanding.

[h0ngxuanli]

Thanks for your efforts! It is true that those fine-grained MD data needs great efforts to be accessible to the public! I am wondering whether there is a way to quickly access only the MD trajectories of all proteins, maybe in a xtc format? 😊

[MihirBafna]

Thanks to the authors for their efforts in releasing this MD dataset!

I'd like to second @h0ngxuanli’s request regarding making the MD trajectories available separately (in .dcd, .xtc, or any other standard trajectory format).

This would significantly reduce the dataset size while still preserving essential dynamic information.
Would it be possible to upload just these files separately?

[byq008]

I truly appreciate your efforts in creating the dynamicPDB dataset. I am particularly interested in the MD data contained within it. I concur with the previous comments and would like to kindly request that you consider uploading the MD trajectories (the DCD files) separately. This would be immensely helpful for further analysis and research.

@h0ngxuanli Hi, thank you for your interest in our dataset.

Before initiating the data upload, we carefully evaluated multiple cloud storage solutions, considering both capacity and bandwidth limitations, and ultimately chose the current platform for its balance of performance and accessibility.

Due to the immense size of the dataset, we are currently able to upload data files for an average of six proteins per day. But we are doing our best to continuously upload the data.

We sincerely appreciate your patience and understanding.