We describe our development of ab initio protein-ligand binding pose prediction models based on transformers and binding affinity prediction models based on the neural tangent kernel (NTK). Folding both protein and ligand, the TwoFold models achieve efficient and quality predictions matching state-of-the-art implementations while additionally reconstructing protein structures. Solving NTK models points to a new use case for highly optimized linear solver benchmarking codes on HPC.
AndoKBaleRLiC, et al. (2022) Digital transformation of droplet/aerosol infection risk assessment realized on “fugaku” for the fight against covid-19. The International Journal of High Performance Computing Applications0(0): 10943420221116056. DOI: 10.1177/10943420221116056
5.
BenigniLPe’che’S (2022) Largest Eigenvalues of the Conjugate Kernel of Single-Layered Neural Networks. arXiv preprint arXiv:2201.04753.
6.
BlanchardAEGounleyJBhowmikD, et al. (2022) Language models for the prediction of sars-cov-2 inhibitors. The International Journal of High Performance Computing Applications0(0): 10943420221121804. DOI: 10.1177/10943420221121804
7.
BohacekRSMcMartinCGuidaWC (1996) The art and practice of structure-based drug design: a molecular modeling perspective. Medicinal Research Reviews16(1): 3–50.
8.
CarsonEHighamNJ (2018) Accelerating the solution of linear systems by iterative refinement in three precisions. SIAM Journal on Scientific Computing40(2): A817–A847.
9.
CasalinoLDommerACGaiebZ, et al. (2021) Ai- driven multiscale simulations illuminate mechanisms of sars- cov-2 spike dynamics. The International Journal of High Performance Computing Applications35(5): 432–451.
10.
DommerACasalinoLKearnsF, et al. (2022) #covidisairborne: Ai-enabled multiscale computational microscopy of delta sars-cov-2 in a respiratory aerosol. The International Journal of High Performance Computing Applications0(0): 1094342022112823. DOI: 10.1177/10943420221128233
11.
ElnaggarAHeinzingerMDallagoCRehawiGWangYJonesLGibbsTFeherTAngererCSteineggerMBHOWMIKDRostB (2020) Prottrans: Towards cracking the language of life’s code through self-supervised deep learning and high performance computing. bioRxiv. DOI: 10.1101/2020.07.12.199554
12.
EvansRO’NeillMPritzelAet al. (2021) Protein complex prediction with alphafold-multimer. bioRxiv.
13.
GaoMNakajimaADParksJM, et al. (2022) Af2complex predicts direct physical interactions in multimeric proteins with deep learning. Nature Communications13(1): 1744.
14.
GlaserJVermaasJVRogersDM, et al. (2021) High-throughput virtual laboratory for drug discovery using massive datasets. The International Journal of High Performance Computing Applications35(5): 452–468.
15.
GlaserJSedovaAGalanieS, et al. (2022) Hit expansion of a noncovalent sars-cov-2 main protease inhibitor. ACS Pharmacology & Translational Science5(4): 255–265.
16.
HastieTMontanariARossetS, et al. (2022) Surprises in high-dimensional ridgeless least squares interpolation. Annals of Statistics50(2): 949.
17.
HekkelmanMLde VriesIJoostenRP, et al. (2021) Alphafill: Enriching the Alphafold Models with Ligands and Co- Factors. bioRxiv.
HronJBahriYSohl-DicksteinJ, et al. (2020) Infinite attention: Nngp and ntk for deep attention networks. In: International Conference on Machine Learning. PMLR, pp. 4376–4386.
20.
JacobsSAMoonTMcLoughlinKet al. (2021) Enabling rapid covid-19 small molecule drug design through scalable deep learning of generative models. The International Journal of High Performance Computing Applications35(5): 469–482.
21.
JacotAGabrielFHonglerC (2018) Neural tangent kernel: convergence and generalization in neural networks. Advances in Neural Information Processing Systems31: 8571–8580.
22.
JumperJEvansRPritzelAet al. (2021a) Applying and improving alphafold at casp14. Proteins: Structure, Function, and Bioinformatics89(12): 1711–1721.
23.
JumperJEvansRPritzelA, et al. (2021b) Highly accurate protein structure prediction with alphafold. Nature596: 583–589. DOI: 10.1038/s41586-021-03819-2
24.
LinZAkinHRaoR, et al. (2022) Language Models of Protein Sequences at the Scale of Evolution Enable Accurate Structure Prediction. bioRxiv.
25.
LuHMathesonMOlesV, et al. (2022a) Climbing the summit and pushing the frontier of mixed precision benchmarks at extreme scale. In: SC22: International Conference for High Performance Computing, Networking, Storage and Analysis.
26.
LuWWuQZhangJ, et al. (2022b) Tankbind: Trigonometry-aware neural networks for drug-protein binding structure prediction. bioRxiv. DOI: 10.1101/2022.06.06.495043
27.
MishraJMishraPAroraNK (2021) Linkages between environmental issues and zoonotic diseases: with reference to covid-19 pandemic. Environmental Sustainability4(3): 455–467.
28.
MuratovENAmaroRAndradeCH, et al. (2021) A Critical Overview of Computational Approaches Employed for Covid-19 Drug Discovery. Chemical Society Reviews.
29.
NovakRXiaoLHronJ, et al. (2019) Neural tangents: Fast and easy infinite neural networks in python. arXiv preprint arXiv:1912.02803.
30.
NovakRSohl-DicksteinJSchoenholzSS (2022) Fast finite width neural tangent kernel. In: International Conference on Machine Learning. PMLR, pp. 17018–17044.
31.
OwenDRAllertonCMAndersonASet al. (2021) An Oral Sars-Cov-2 Mpro Inhibitor Clinical Candidate forthe Treatment of covid-19. Science374(6575): 1586–1593.
32.
PaulSMMytelkaDSDunwiddieCT, et al. (2010) How to Improve R and d.
33.
PooleBLahiriSRaghuM, et al. (2016) Exponential expressivity in deep neural networks through transient chaos. Advances in Neural Information Processing Systems29.
34.
RabeMNStaatsC (2021) Self-attention does not need o(n2) memory DOI:10.48550/ARXIV.2112.05682
ReymondJL (2015) The chemical space project. Accounts of Chem- ical Research48(3): 722–730. DOI: 10.1021/ar500432k
37.
Ruiz EstradaMAKhanA (2020) Globalization and pandemics: the case of covid-19. Available at SSRN 3560681.
38.
SchwallerPGaudinTLa´nyiD, et al. (2018) “Found in translation”: predicting outcomes of complex organic chemistry reactions using neural sequence-to-sequence models. Chemical Science9: 6091–6098.
39.
ShazeerNSternM (2018) Adafactor: adaptive learning rates with sublinear memory cost. In: International Conference on Machine Learning. PMLR, pp. 4596–4604.
40.
ShenTHuZPengZ, et al. (2022) E2efold-3d: end-to- end deep learning method for accurate de novo rna 3d structure prediction. arXiv preprint arXiv:2207.01586.
41.
StärkHGaneaOEPattanaikL, et al. (2022) Equibind: Geometric deep learning for drug binding structure prediction. arXiv preprint arXiv:2022.05146.
42.
StrangG (2016) Introduction to Linear Algebra. 5th edn.Welles-ley, MA: Wellesley-Cambridge Press. ISBN 9780980232776.
43.
SuMYangQDuY, et al. (2018) Comparative assessment of scoring functions: the casf-2016 update. Journal of Chemical Information and Modeling59(2): 895–913.
VaradiMAnyangoSDeshpandeM, et al. (2022) AlphaFold protein structure database: massively expanding the structural coverage of protein-sequence space with high-accuracy models. Nucleic acids research50(D1): D439–D444.
46.
VaswaniAShazeerNParmarN, et al. (2017) Attention is all you need. Advances in Neural Information Processing Systems30.
47.
WHO (2022) WHO Coronavirus (COVID-19) Dashboard.
48.
XiaoLPenningtonJSchoenholzS (2020) Disentangling trainability and generalization in deep neural networks. In: International Conference on Machine Learning. PMLR, pp. 10462–10472.
49.
XiaoLHuHMisiakiewiczT, et al. (2022) Precise learning curves and higher-order scalings for dot- product kernel regression. Advances in Neural Information Processing Systems35: 4558–4570.
50.
ZhongBSuXWenM, et al. (2021) Parafold: Paralleling alphafold for large-scale predictions.
