Molecular Structure Prediction

The acquisition of molecular structure, especially the structure of biomacromolecules (DNA, RNA and protein), is an important issue in the field of biopharmaceutical research and has a wide range of uses. The MindSpore SPONGE biological computing toolkit provides a series of calculation tools for molecular structure prediction, helping researchers to acquire high-precision molecular structure efficiently.

Currently, a series of tools for protein and RNA structure prediction are available, which support high precision prediction of protein monomer and complex structure and RNA secondary structure.

Supported Networks

Function	Model	Training	Inferring	Back-end
Single Chain Structure Prediction	MEGA-Fold	√	√	GPU/Ascend
MSA Generation/Correction	MEGA-EvoGen	√	√	GPU/Ascend
Structural Quality Assessment	MEGA-Assessment	√	√	GPU/Ascend
Multi-chain Structure Prediction	AlphaFold-Multimer	×	√	GPU/Ascend
RNA Secondary Structure Prediction	UFold	√	√	GPU/Ascend

In the future, we will further improve the function of molecular structure prediction, and introduce more tools for protein-ligand complex structure prediction and small molecule structure prediction of compounds. Stay tuned for more information.