Enhancing AlphaFold-Multimer-based Protein Complex Structure Prediction with MULTICOM in CASP15

Liu, Jian; Guo, Zhiye; Wu, Tai Tsun; Roy, Raj; Quadir, Farhan; Chen, Chen; Cheng, Jianlin

doi:10.1101/2023.05.16.541055

Cited by 12 publications

(8 citation statements)

References 36 publications

(26 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other ways of improving MSA usage by AFM have also been proposed [63] and could be combined with advances in pairing. Besides improving MSA construction [64] and the extraction of MSA information, other promising approaches include exploiting structural alignments [65], using massive sampling and dropout [66], and combining AFM with more traditional docking methods [67,68], which has allowed e.g. to improve structure prediction of 6A6I [67].…”

Section: Discussionmentioning

confidence: 99%

Pairing interacting protein sequences using masked language modeling

Lupo

Sgarbossa

Bitbol

2023

Preprint

View full text Add to dashboard Cite

Predicting which proteins interact together from amino-acid sequences is an important task. We develop a method to pair interacting protein sequences which leverages the power of protein language models trained on multiple sequence alignments, such as MSA Transformer and the EvoFormer module of AlphaFold. We formulate the problem of pairing interacting partners among the paralogs of two protein families in a differentiable way. We introduce a method called DiffPALM that solves it by exploiting the ability of MSA Transformer to fill in masked amino acids in multiple sequence alignments using the surrounding context. MSA Transformer encodes coevolution between functionally or structurally coupled amino acids. We show that it captures inter-chain coevolution, while it was trained on single-chain data, which means that it can be used out-of-distribution. Relying on MSA Transformer without fine-tuning, DiffPALM outperforms existing coevolution-based pairing methods on difficult benchmarks of shallow multiple sequence alignments extracted from ubiquitous prokaryotic protein datasets. It also outperforms an alternative method based on a state-of-the-art protein language model trained on single sequences. Paired alignments of interacting protein sequences are a crucial ingredient of supervised deep learning methods to predict the three-dimensional structure of protein complexes. DiffPALM substantially improves the structure prediction of some eukaryotic protein complexes by AlphaFold-Multimer, without significantly deteriorating any of those we tested. It also achieves competitive performance with using orthology-based pairing.

show abstract

Section: Discussionmentioning

confidence: 99%

Pairing interacting protein sequences using masked language modeling

Lupo

Sgarbossa

Bitbol

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Advanced ML techniques are being applied to rapidly solve structural biology problems [55][56][57]. By reviewing the literature, we found that ML is widely used in methods to predict the impact of GV on protein structure.…”

Section: Based On Machine Learningmentioning

confidence: 99%

Effects of genetic variation on the structure of RNA and protein

Kang,

Wei,

Jia

et al. 2024

Proteomics

View full text Add to dashboard Cite

Changes in the structure of RNA and protein, have an important impact on biological functions and are even important determinants of disease pathogenesis and treatment. Some genetic variations, including copy number variation, single nucleotide variation, and so on, can lead to changes in biological function and increased susceptibility to certain diseases by changing the structure of RNA or protein. With the development of structural biology and sequencing technology, a large amount of RNA and protein structure data and genetic variation data resources has emerged to be used to explain biological processes. Here, we reviewed the effects of genetic variation on the structure of RNAs and proteins, and investigated their impact on several diseases. An online resource (http://www.onethird‐lab.com/gems/) to support convenient retrieval of common tools is also built. Finally, the challenges and future development of the effects of genetic variation on RNA and protein were discussed.

show abstract

“…AlphaFold-multimer predictions--We used Alphafold-multimer v2.3.1, 3D models of the human LRRK2-RAB10 complex. The necessary databases for running AlphaFold-multimer were downloaded on 12 January 2023 (Liu et al, 2023).…”

Section: Analysis Of the Model Ensemble--mentioning

confidence: 99%

Biophysical analysis reveals autophosphorylation as an important negative regulator of LRRK2 dimerization

Guaitoli,

Zhang,

Saitta

et al. 2023

Preprint

View full text Add to dashboard Cite

Leucine-rich repeat kinase 2 (LRRK2) is a large, multi-domain protein which is associated with Parkinson's disease. Although high-resolution structures of LRRK2 are available, little is known about the complex dynamics behind the inter-domain regulation of LRRK2 and its perturbation by pathogenic variants. Previous studies have demonstrated that LRRK2 goes through an oligomerization cycle at the membrane, however it remains unclear in which form it exerts its kinase activity. Moreover, the LRRK2 monomer-dimer equilibrium and associated functional implications at a molecular level also need further investigation. In the present work, we used a multi-faceted approach to better understand LRRK2 oligomerization and suggest a functional model of how LRRK2 interacts with its substrates. To this end, we combined nano differential scanning calorimetry and mass photometry with molecular modelling. The thermal analysis resulted in a multistep denaturation profile, elucidating novel insights into the composite structural organization of the multi-domain protein LRRK2. Furthermore, LRRK2 shows a remarkable thermal stability, confirming its oligomeric nature. By using mass photometry, we could observe a monomer-dimer equilibrium which is altered by R1441G, a pathogenic variant within the Roc-COR interface. Most importantly, we could demonstrate that autophosphorylation induces LRRK2 monomerization, indicating a novel intramolecular feedback mechanism. Finally, we investigated the interaction of LRRK2 with its substrate, RAB10 by integrative computational modelling. The resulting models suggest that the monomeric form of LRRK2 is the favored protein conformation for the interaction with its substrate, leading to an increasing interest in the monomer-dimer equilibrium as a possible intervention point for the pathology.

show abstract

Enhancing AlphaFold-Multimer-based Protein Complex Structure Prediction with MULTICOM in CASP15

Cited by 12 publications

References 36 publications

Pairing interacting protein sequences using masked language modeling

Pairing interacting protein sequences using masked language modeling

Effects of genetic variation on the structure of RNA and protein

Biophysical analysis reveals autophosphorylation as an important negative regulator of LRRK2 dimerization

Contact Info

Product

Resources

About