Thank you very much for creating such a great tool for predicting protein-DNA interactions! I have been using this program for over a year, starting before the release of version 0.2. I encountered some inaccuracies in structures published after April 30, 2020 (the training cut off indicated in biorxiv paper). However, with the release of version 0.2, the performance has significantly improved. Could you please clarify if these improvements are solely due to the new version of the program? Also, I am interested to know if the training cut-off is consistent across all versions.
Dear RosettaFoldNA Developers,
Thank you very much for creating such a great tool for predicting protein-DNA interactions! I have been using this program for over a year, starting before the release of version 0.2. I encountered some inaccuracies in structures published after April 30, 2020 (the training cut off indicated in biorxiv paper). However, with the release of version 0.2, the performance has significantly improved. Could you please clarify if these improvements are solely due to the new version of the program? Also, I am interested to know if the training cut-off is consistent across all versions.
Thank you very much in advance!
Best wishes, Ilya