ParaFold: Paralleling AlphaFold for Large-Scale Predictions

Zhong, Bozitao; Su, Xiaoming; Wen, Minhua; Zuo, Sichen; Hong, Liang; Lin, James

doi:10.1145/3503470.3503471

Cited by 22 publications

(19 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The PREDDIMER webserver was used both to predict models for the 11 TM dimers ab initio , but also to analyse the NMR and CG MD derived structures based on the Fscor and helix crossing angle (Polyansky et al, 2014b). Alphafold2 Multimer (Evans et al, 2022) was used to predict dimer models for the same set of complexes, using a local installation of the ParaFold pipeline (Zhong et al, 2022). This installation was recently validated by Martin and colleagues in 2022 (Martin, 2022), reproducing the results predicted by Evans et al, 2021 (Evans et al, 2022).…”

Section: Discussionmentioning

confidence: 99%

Transmembrane region dimer structures of Type 1 receptors readily sample alternate configurations: MD simulations using the Martini 3 coarse grained model compared to AlphaFold2 Multimer

Sahoo

Souza

Meng

et al. 2021

Preprint

View full text Add to dashboard Cite

Determination of the structure and dynamics of transmembrane (TM) domains of single-transmembrane receptors is key to understanding their mechanism of signal transduction across the plasma membrane. Although many studies have been performed on isolated soluble extra- and intracellular receptor domains in aqueous solutions, limited knowledge exists on the lipid embedded TM region. In this study, we predict the assembly of alternate configurations of receptor TM domain dimers using the Martini 3 force field for coarse-grain (CG) molecular dynamic simulations. This recent version of Martini has new bead types and sizes, which allows more accurate predictions of molecular packing and interactions compared to the previous version. Our results with Martini 3 simulations show overall good agreement with ab initio predictions using PREDDIMER and with available NMR derived structures for TM helix dimers. Understanding and predicting the association TM domains may help us to better understand the signalling mechanism of TM receptors, in turn providing the opportunity for development of new pharmaceuticals, some of which are peptide based.

show abstract

Section: Discussionmentioning

confidence: 99%

Transmembrane region dimer structures of Type 1 receptors readily sample alternate configurations: MD simulations using the Martini 3 coarse grained model compared to AlphaFold2 Multimer

Sahoo

Souza

Meng

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Despite a great diversity of peptide features (classical and non-classical) that has been used in AMPs prediction/design, most of those features are sequences- or property- based; however, the 3D structural information of AMPs has not been deeply exploited for such aims [ 173 , 174 , 175 ]. Although experimental determinate 3D structures of AMPs are used in minor proportion than their sequences, the 3D structure prediction tools are becoming more accessible and less computational demanding when considering new advances in both software and computer architectures [ 176 , 177 ]. These facts will ease the gradually inclusion of 3D structural features in the prediction models.…”

Section: Future Research Directionsmentioning

confidence: 99%

Emerging Computational Approaches for Antimicrobial Peptide Discovery

et al. 2022

View full text Add to dashboard Cite

In the last two decades many reports have addressed the application of artificial intelligence (AI) in the search and design of antimicrobial peptides (AMPs). AI has been represented by machine learning (ML) algorithms that use sequence-based features for the discovery of new peptidic scaffolds with promising biological activity. From AI perspective, evolutionary algorithms have been also applied to the rational generation of peptide libraries aimed at the optimization/design of AMPs. However, the literature has scarcely dedicated to other emerging non-conventional in silico approaches for the search/design of such bioactive peptides. Thus, the first motivation here is to bring up some non-standard peptide features that have been used to build classical ML predictive models. Secondly, it is valuable to highlight emerging ML algorithms and alternative computational tools to predict/design AMPs as well as to explore their chemical space. Another point worthy of mention is the recent application of evolutionary algorithms that actually simulate sequence evolution to both the generation of diversity-oriented peptide libraries and the optimization of hit peptides. Last but not least, included here some new considerations in proteogenomic analyses currently incorporated into the computational workflow for unravelling AMPs in natural sources.

show abstract

“…Fortunately, its open source has attracted joint efforts of developers to improve it. In 2021, Zhong and colleagues, from the Center for High Performance Computing (HPC) at Shanghai Jiao Tong University, released ParaFold [ 21 ], the specific AlphaFold version for their HPC clusters, making a successful attempt to accelerate AlphaFold2. In May 2022, ColabFold [ 22 ], which claimed to make protein folding accessible to all, was officially published.…”

Section: Introductionmentioning

confidence: 99%

Utilizing an artificial intelligence system to build the digital structural proteome of reef-building corals

et al. 2022

View full text Add to dashboard Cite

Background Reef-building corals play an important role in the marine ecosystem, and analyzing their proteomes from a structural perspective will exert positive effects on exploring their biology. Here we integrated mass spectrometry with newly published ColabFold to obtain digital structural proteomes of dominant reef-building corals. Results Of the 8,382 homologous proteins in Acropora muricata, Montipora foliosa, and Pocillopora verrucosa identified, 8,166 received predicted structures after about 4,060 GPU hours of computation. The resulting dataset covers 83.6% of residues with a confident prediction, while 25.9% have very high confidence. Conclusions Our work provides insight-worthy predictions for coral research, confirms the reliability of ColabFold in practice, and is expected to be a reference case in the impending high-throughput era of structural proteomics.

show abstract

ParaFold: Paralleling AlphaFold for Large-Scale Predictions

Cited by 22 publications

References 9 publications

Transmembrane region dimer structures of Type 1 receptors readily sample alternate configurations: MD simulations using the Martini 3 coarse grained model compared to AlphaFold2 Multimer

Transmembrane region dimer structures of Type 1 receptors readily sample alternate configurations: MD simulations using the Martini 3 coarse grained model compared to AlphaFold2 Multimer

Emerging Computational Approaches for Antimicrobial Peptide Discovery

Utilizing an artificial intelligence system to build the digital structural proteome of reef-building corals

Contact Info

Product

Resources

About