Prof. White received a grant from the National Institute of Health (NIH). The project is titled "Learning to learn in structural biology with deep neural networks." The goal of this research is to develop methods that allow meta-learning
in structural biology to train new deep learning models on tasks where data is scarce, with limited computating resources.
May - Aug, 2020
Congratulations to Dilnoza, Maghesree and Rainier for defending their PhD thesis! Dr. Dilnoza Amirkulova joined PnG as a Scientist, Dr. Maghesree Chakraborty joined Intel as a RET Design Engineer, and Dr. Rainier Barrett joined
the Jankowski lab at Boise State University as a post-doctoral researcher.
Geemi received a COVID-19 Seed fellowship from the Molecular Sciences Software Institute (MolSSI) to study the SARS-CoV-2 protease using coarse-graining. Heta received a Software Investment fellowship from MolSSI to develop HOOMD-TF
for coarse-grained simulations in January 2020. Congratulations!
Prof. White and co-PI Gourab Ghoshal received a grant from the National Science Foundation. The goal of this project is to use maximum entropy biasing to create accurate COVID-19 models with sufficient complexity to inform local
The White lab hosted a four-day workshop for high school students where they a chance to learn about molecular chemistry and visualize simulations in VR. This workshop was funded by the NSF and took place as a part of the Upward
Bound program at the UofR.
May 23, 2019
Maghesree won the first prize for her lightening talk at the RAMP 2019 Symposium. Congratulations Maghesree! Earlier this year, in April, she won the best poster award at the Chemistry Chemical Engineering Poster Social organized
by the Chemistry Graduate Association.
March 26, 2019
Prof. White was awarded the G. Graydon Curtis '58 and Jane W. Curtis Award for Nontenured Faculty Teaching.
Abstract for Graph Neural Network Based Coarse-Grained Mapping Prediction:
The selection of coarse-grained (CG) mapping operators is a critical step for CG molecular dynamics (MD) simulation. It is still an open question about what is optimal for this choice and there is a need for theory. The current state-of-the art method
is mapping operators manually selected by experts. In this work, we demonstrate an automated approach by viewing this problem as supervised learning where we seek to reproduce the mapping operators produced by experts. We present
a graph neural network based CG mapping predictor called DEEP SUPERVISED GRAPH PARTITIONING MODEL (DSGPM) that treats mapping operators as a graph segmentation problem. DSGPM is trained on a novel dataset, Human-annotated Mappings
(HAM), consisting of 1,206 molecules with expert annotated mapping operators. HAM can be used to facilitate further research in this area. Our model uses a novel metric learning objective to produce high-quality atomic features
that are used in spectral clustering. The results show that the DSGPM outperforms state-of-the-art methods in the field of graph segmentation.