We just released a new update for MAJIQ 2 branch. MAJIQ 2.2 has a number of improvements and fixes in many edge cases during the majiq build process, as well as incremental build mode. In addition, there are small performance improvements, in majiq and voila and other new minor features.
Please see the full patch nodes and download details at the pinned thread here.
Thanks for your help and we hope you enjoy the new majiq, great new features are coming soon.
MAJIQ 2.0 has been released. It represents a significant upgrade compared to previous MAJIQ 1.1. Main advantages and changes include:
A significant gain in speed. This is the result of algorithmic improvements and porting much of the core to C++.
New algorithm for intron retention. The new algorithm is both faster and more accurate.
New visualization engine: The release includes VOILA 2.0 which is now run as local service. This allows for smoother/faster interaction and enhanced features such as a built in search option.
Output a negative set: When comparing two conditions MAJIQ can give not only a high confidence set of changing events but also a high confidence set of events that are *not* changing.
As a consequence of these changes, MAJIQ 2.0 can easily handle hundreds and even thousands of samples. Most importantly, 2.0 serves as a base for many exciting new features and algorithms that support new use cases, so stay tuned!
Finally, while the academic license remains the same please note that commercial license has changed. Commercial users can keep using 1.1 as before, if they want to upgrade to MAJIQ 2.0 they should follow the instructions found on https://majiq.biociphers.org/