Big news! PMN has just received a new grant from the National Science Foundation's Plant Genome Research Program (PGRP). Worth $3 million over four years, this grant will allow us to significantly expand PMN in size, scope, and usefulness. In addition to our PI, Dr. Sue Rhee here at Michigan State University, whose lab created PMN and has operated and maintained it for more than 15 years, we have now also brought on collaborators Dr. Hiroshi Maeda at the University of Wisconsin-Madison, and Dr. Phillipp Zerbe from the University of California, Davis. Both these labs have tremendous expertise in the functional characterization of enzymes, and they have signed on to work with us to discover new enzymes as well as new functions for previously-known enzymes in a multitude of plant species. Newly-characterized enzymes will be represented in PMN directly in the appropriate species database, but will also be included in the dataset we use to make computational predictions of enzymes in all PMN species, so the impact of these new data will reach beyond their species of origin. These labs will form the core of an Enzyme Consortium, a collaboration of computational and experimental groups focused on identifying and filling gaps in our understanding of plant enzymes and metabolism.
As another major objective, we plan to massively increase the number of plant and green alge species databases in PMN, from the current total of 127 to more than 1000 by the end of the grant period. To achieve this increase, we will be overhauling the computational pipeline we use to create and update the PMN databases, making it faster, easier, more reproducible, and more scalable. Even better, we will be releasing the upgraded pipeline publicly as a Singularity container, so interested users can reproduce our results and generate their own plant databases comparable to the ones in PMN. Our next release, PMN 16, is due out in mid-2024 and will be the first to be built using the new pipeline.
We will also be releasing a new version of our Ensemble Enzyme Prediction Pipeline (E2P2) software [insert link to github], a core piece of the pipeline that predicts enzyme function from amino acid sequences. E2P2 is an ensemble classifier, meaning it combines the predictions of different classifier software to produce a final set of predictions, and the new E2P2 version 5 will be modularized so that users can easily add and swap out classifiers using a standardized interface. The default classifiers will be blastp [link to ncbi blast+] and DeepEC [link to paper], the latter replacing PRIAM which is no longer maintained, but users will be able to easily add more if desired.
Finally, we will be developing educational resources and materials to help K-12 teachers engage students from diverse backgrounds with plant science, helping to fostering the next generation of leaders in the field.
We're super excited to take PMN to the next level, and we can't wait to get started!
