English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Accurate protein structure prediction can significantly accelerate thedevelopment of life science. The accuracy of AlphaFold2, a frontier end-to-endstructure prediction system, is already close to that of the experimentaldetermination techniques. Due to the complex model architecture and largememory consumption, it requires lots of computational resources and time toimplement the training and inference of AlphaFold2 from scratch. The cost ofrunning the original AlphaFold2 is expensive for most individuals andinstitutions. Therefore, reducing this cost could accelerate the development oflife science. We implement AlphaFold2 using PaddlePaddle, namely HelixFold, toimprove training and inference speed and reduce memory consumption. Theperformance is improved by operator fusion, tensor fusion, and hybridparallelism computation, while the memory is optimized through Recompute,BFloat16, and memory read/write in-place. Compared with the original AlphaFold2(implemented with Jax) and OpenFold (implemented with PyTorch), HelixFold needsonly 7.5 days to complete the full end-to-end training and only 5.3 days whenusing hybrid parallelism, while both AlphaFold2 and OpenFold take about 11days. HelixFold saves 1x training time. We verified that HelixFold's accuracycould be on par with AlphaFold2 on the CASP14 and CAMEO datasets. HelixFold'scode is available on GitHub for free download:https://github.com/PaddlePaddle/PaddleHelix/tree/dev/apps/protein_folding/helixfold,and we also provide stable web services onhttps://paddlehelix.baidu.com/app/drug/protein/forecast.

HelixFold: An Efficient Implementation of AlphaFold2 using PaddlePaddle