The benchmark dataset contains 272 mycobacterial protein sequences. These protein sequences share the sequence identity of <25%.

Download