How to interpret the results

[szblh]

Hi,
Thanks for your excellent work. I am using this package to derive some potential targets right now. Using the test dataset, I can get result of two files: auc.tsv and probs.tsv. I am wondering how to interpret the results. It will be helpful if you can provide me some information.

[ozlemmuslu]

Hi, There are two files generated by the program: 1. auc.tsv gives you the performance on repeated cross validation, you can use this to understand how good the predictions were for your use case 2. probs.tsv gives you the probability of being a target for each of the candidates that is both in your network and your differential gene analysis. If you sort them, you can find the highest rank candidates. You might want to ignore already known candidates here, and I'd recommend exploring top candidates (e.g. looking them up in the literature) to identify a suitable one for your purpose. The tool is assigning scores based on similarity/interaction, so further analysis is necessary. Hope this helps

…

On Thu, Mar 26, 2020, 03:41 szblh ***@***.***> wrote: Hi, Thanks for your excellent work. I am using this package to derive some potential targets right now. Using the test dataset, I can get result of two files: auc.tsv and probs.tsv. I am wondering how to interpret the results. It will be helpful if you can provide me some information. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#2>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABFKXM2AAEE6YPCM5DXP5ALRJK6EZANCNFSM4LT44JQQ> .