Will you upload Supplementary Section ?

[Kaiton7]

Hello. In the paper(https://www.biorxiv.org/content/10.1101/642926v3.full), "We simulate these networks using BoolODE, a method we have developed to convert Boolean models into systems of differential equations (Supplementary Section S1)", but I can't see the Supplementary Section1. So, I cant't know how to create the input data for synthetic network. I'd like to use the synthetic data which used in this paper. Could you upload Supplementary Section1 and Table2 ? Thank you.

[adyprat]

Hi, Thank you for your interest in our paper. The supplementary section S1 and tables are already available on bioRxiv in the pdf version of the paper (pg. 22-24, and pg. 41 here: https://www.biorxiv.org/content/early/2019/06/04/642926.full.pdf).

Note that the 'Full Text' version on bioRxiv is automatically generated, so it may have skipped the supplementary sections while converting from pdf.

We have since also uploaded the simulated datasets used in BEELINE on zenodo.org, and they are now accessible here: https://zenodo.org/record/3378976 .

The BoolODE commands used to generate these datasets is available at https://github.com/Murali-group/BoolODE/blob/master/scripts/script-to-generate-datasets.sh. The documentation for BoolODE is available at https://murali-group.github.io/Beeline/BoolODE.html.

Let us know if you have any further questions about the datasets used in BEELINE.

Thank you, Aditya

[Kaiton7]

Dear Aditya Thank you so much. I see now. And I think I've learn a lot. In this page:(https://zenodo.org/record/3378976), there are dyn-LL-100-1, dyn-LL-100-2,... dyn-LL-100-10. I think each of files are independently and each of them can be used for evaluation independently. So, which file did you use for evaluation in your paper?? Thank you. Kaito

[adyprat]

Hi Kaito,

In the bioRxiv version we reported the results for datasets with 500, 2000 and 5000 cells for Synthetic Networks. We have since added datasets with 100 and 200 cells as well. The README.txt in the zenodo .zip download has more details on folder names.

You are correct in thinking that the evaluation is done independently for dyn-LL-100-1, dyn-LL-100-2, and so on. We reported the median AUPRC and AUROC in the paper, i.e., median AUPRC and AUROCs, obtained across each of:

• 10 datasets with 500 cells ( dyn-LL-500-1, dyn-LL-500-2,... dyn-LL-500-10)
• 10 datasets with 2000 cells (dyn-LL-2000-1, dyn-LL-2000-2,... dyn-LL-2000-10), and
• 10 datasets with 5000 cells (dyn-LL-5000-1, dyn-LL-5000-2,... dyn-LL-5000-10)

Hope that clarifies your doubt. Let me know if you have any further questions.

Thanks, Aditya

[Kaiton7]

Thank you so much for your prompt reply. I understand it. I'm quite interested in this experiments. Thank you. Best Regards Kaito