Thanks in major part to the amazing work of Harvey Feng, a distributed, batch version of eXpress can now be run on a cluster using Apache Spark to provide a scalable solution for fragment assignment and abundance estimation. Since eXpress-D uses the full batch EM algorithm, it provides the most accurate estimates according to our tests.
For more details, please read our manuscript in BMC Bioinformatics and check out the wiki on GitHub.
This update is a minor patch containing the following bug fixes: