Modelling protein complexes with crosslinking mass spectrometry and deep learning

Stahl, Kolja; Brock, Oliver; Rappsilber, Juri

doi:10.1101/2023.06.07.544059

Cited by 17 publications

(13 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The AlphaLink (40) prediction (model confidence: 0.75) of the B. subtilis Fur-dimer (Fig. 6A) resembles a V-shaped conformation, similar to other Fur-proteins which interact with DNA (41).…”

Section: Resultsmentioning

confidence: 72%

See 1 more Smart Citation

Control of iron homeostasis by a regulatory protein-protein interaction inBacillus subtilis: The FurA (YlaN) acts as an antirepressor to the ferric uptake regulator Fur

Demann,

Bremenkamp,

Stahl

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

Iron is essential for most organisms. However, two problems are associated with the use of iron for aerobically growing organisms: (i) its accumulation leads to the formation of toxic reactive oxygen species and (ii) it is present mainly as the highly insoluble ferric iron which makes the access to iron difficult. As a consequence, a tight regulation of iron homeostasis is required. This regulation is achieved in many bacteria by the ferric uptake repressor Fur. The way how the activity of Fur is controlled, has so far remained elusive. Here, we have identified the Fur antirepressor FurA (previously YlaN) in the model bacteriumBacillus subtilisand describe its function to release Fur from the DNA under conditions of iron limitation. The FurA protein physically interacts with Fur, and this interaction prevents Fur from binding to its target sites due to a complete re-orientation of the protein. Bothin vivoandin vitroexperiments using a reporter fusion and Fur-DNA binding assays, respectively, demonstrate that the Fur-FurA interaction prevents Fur from binding DNA and thus from repressing the genes required for iron uptake. Accordingly, the lack of FurA results in the inability of the cell to express the genes for iron uptake under iron-limiting conditions. This explains why thefurAgene was identified as being essential under standard growth conditions inB. subtilis. Phylogenetic analysis suggests that the control of Fur activity by the antirepressor FurA is confined to, but very widespread in bacteria of the class Bacilli.IMPORTANCEIron is essential for most bacteria since it is required for many redox reactions. Under aerobic conditions, iron is both essential and toxic due to radical formation. Thus, iron homeostasis must be faithfully controlled. The transcription factor Fur is responsible for this regulation in many bacteria; however, the control of Fur activity has remained open. Here we describe the FurA protein, a so far unknown protein which acts as an antirepressor to Fur inBacillus subtilis. This mechanism seems to be widespread inB. subtilisand several important pathogens and might be a promising target for drug development.

show abstract

Section: Resultsmentioning

confidence: 72%

“…We predicted the FurA-Fur complex structure with AlphaLink2 v2 (40) using crosslinking MS data (34). For each prediction, we ran 3 recycling iterations.…”

Section: Methodsmentioning

confidence: 99%

Control of iron homeostasis by a regulatory protein-protein interaction inBacillus subtilis: The FurA (YlaN) acts as an antirepressor to the ferric uptake regulator Fur

Demann,

Bremenkamp,

Stahl

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The first one, AlphaLink, incorporates experimental distances obtained from cross-linking mass spectrometry to guide AF2M-based complex formation. 89,90 The second one, RoseTTAFold2, leverages the strengths of both RoseTTAFold and AF2 and has shown scalability in modeling large complexes. 91 Additionally, the success of CASP15 has motivated DeepMind to release an updated version of AF2, known as AF2-v2.3.0.…”

Section: Discussionmentioning

confidence: 99%

The impact of AI‐based modeling on the accuracy of protein assembly prediction: Insights from CASP15

Ozden,

Kryshtafovych,

Karaca

2023

Proteins

View full text Add to dashboard Cite

In CASP15, 87 predictors submitted around 11 000 models on 41 assembly targets. The community demonstrated exceptional performance in overall fold and interface contact predictions, achieving an impressive success rate of 90% (compared to 31% in CASP14). This remarkable accomplishment is largely due to the incorporation of DeepMind's AF2‐Multimer approach into custom‐built prediction pipelines. To evaluate the added value of participating methods, we compared the community models to the baseline AF2‐Multimer predictor. In over 1/3 of cases, the community models were superior to the baseline predictor. The main reasons for this improved performance were the use of custom‐built multiple sequence alignments, optimized AF2‐Multimer sampling, and the manual assembly of AF2‐Multimer‐built subcomplexes. The best three groups, in order, are Zheng, Venclovas, and Wallner. Zheng and Venclovas reached a 73.2% success rate over all (41) cases, while Wallner attained 69.4% success rate over 36 cases. Nonetheless, challenges remain in predicting structures with weak evolutionary signals, such as nanobody–antigen, antibody–antigen, and viral complexes. Expectedly, modeling large complexes also remains challenging due to their high memory compute demands. In addition to the assembly category, we assessed the accuracy of modeling interdomain interfaces in the tertiary structure prediction targets. Models on seven targets featuring 17 unique interfaces were analyzed. Best predictors achieved a 76.5% success rate, with the UM‐TBM group being the leader. In the interdomain category, we observed that the predictors faced challenges, as in the case of the assembly category, when the evolutionary signal for a given domain pair was weak or the structure was large. Overall, CASP15 witnessed unprecedented improvement in interface modeling, reflecting the AI revolution seen in CASP14.

show abstract

“…Mass spectrometry (MS) is another experimental approach which well complements structure prediction methods. Indeed, several studies have shown how MS data can be combined with AF2 models. − Specific tools have been developed to include MS data in the modeling process, namely, AlphaLink and its extension to AlphaFold-Multimer, AlphaLink2 …”

Section: Integrating Af2 Into Experimental Pipelinesmentioning

confidence: 99%

A Perspective on the Prospective Use of AI in Protein Structure Prediction

Versini,

Sritharan,

Aykac Fas

et al. 2023

J. Chem. Inf. Model.

View full text Add to dashboard Cite

AlphaFold2 (AF2) and RoseTTaFold (RF) have revolutionized structural biology, serving as highly reliable and effective methods for predicting protein structures. This article explores their impact and limitations, focusing on their integration into experimental pipelines and their application in diverse protein classes, including membrane proteins, intrinsically disordered proteins (IDPs), and oligomers. In experimental pipelines, AF2 models help X-ray crystallography in resolving the phase problem, while complementarity with mass spectrometry and NMR data enhances structure determination and protein flexibility prediction. Predicting the structure of membrane proteins remains challenging for both AF2 and RF due to difficulties in capturing conformational ensembles and interactions with the membrane. Improvements in incorporating membrane-specific features and predicting the structural effect of mutations are crucial. For intrinsically disordered proteins, AF2's confidence score (pLDDT) serves as a competitive disorder predictor, but integrative approaches including molecular dynamics (MD) simulations or hydrophobic cluster analyses are advocated for accurate dynamics representation. AF2 and RF show promising results for oligomeric models, outperforming traditional docking methods, with AlphaFold-Multimer showing improved performance. However, some caveats remain in particular for membrane proteins. Real-life examples demonstrate AF2's predictive capabilities in unknown protein structures, but models should be evaluated for their agreement with experimental data. Furthermore, AF2 models can be used complementarily with MD simulations. In this Perspective, we propose a "wish list" for improving deep-learning-based protein folding prediction models, including using experimental data as constraints and modifying models with binding partners or post-translational modifications. Additionally, a meta-tool for ranking and suggesting composite models is suggested, driving future advancements in this rapidly evolving field.

show abstract

Modelling protein complexes with crosslinking mass spectrometry and deep learning

Cited by 17 publications

References 42 publications

Control of iron homeostasis by a regulatory protein-protein interaction inBacillus subtilis: The FurA (YlaN) acts as an antirepressor to the ferric uptake regulator Fur

Control of iron homeostasis by a regulatory protein-protein interaction inBacillus subtilis: The FurA (YlaN) acts as an antirepressor to the ferric uptake regulator Fur

The impact of AI‐based modeling on the accuracy of protein assembly prediction: Insights from CASP15

A Perspective on the Prospective Use of AI in Protein Structure Prediction

Contact Info

Product

Resources

About