An official website of the United States government
Abstract MotivationQuality assessment (QA) of predicted protein tertiary structure models plays an important role in ranking and using them. With the recent development of deep learning end-to-end protein structure prediction techniques for generating highly confident tertiary structures for most proteins, it is important to explore corresponding QA strategies to evaluate and select the structural models predicted by them since these models have better quality and different properties than the models predicted by traditional tertiary structure prediction methods. ResultsWe develop EnQA, a novel graph-based 3D-equivariant neural network method that is equivariant to rotation and translation of 3D objects to estimate the accuracy of protein structural models by leveraging the structural features acquired from the state-of-the-art tertiary structure prediction method—AlphaFold2. We train and test the method on both traditional model datasets (e.g. the datasets of the Critical Assessment of Techniques for Protein Structure Prediction) and a new dataset of high-quality structural models predicted only by AlphaFold2 for the proteins whose experimental structures were released recently. Our approach achieves state-of-the-art performance on protein structural models predicted by both traditional protein structure prediction methods and the latest end-to-end deep learning method—AlphaFold2. It performs even better than the model QA scores provided by AlphaFold2 itself. The results illustrate that the 3D-equivariant graph neural network is a promising approach to the evaluation of protein structural models. Integrating AlphaFold2 features with other complementary sequence and structural features is important for improving protein model QA. Availability and implementationThe source code is available at https://github.com/BioinfoMachineLearning/EnQA. Supplementary informationSupplementary data are available at Bioinformatics online.
more »
« less
Neural network conditioned to produce thermophilic protein sequences can increase thermal stability
Abstract This work presents Neural Optimization for Melting-temperature Enabled by Leveraging Translation (NOMELT), a novel approach for designing and ranking high-temperature stable proteins using neural machine translation. The model, trained on over 4 million protein homologous pairs from organisms adapted to different temperatures, demonstrates promising capability in targeting thermal stability. A designed variant of theDrosophila melanogasterEngrailed Homeodomain shows a melting temperature increase of 15.5 K. Furthermore, NOMELT achieves zero-shot predictive capabilities in ranking experimental melting and half-activation temperatures across a number of protein families. It achieves this without requiring extensive homology data or massive training datasets as do existing zero-shot predictors by specifically learning thermophilicity, as opposed to all natural variation. These findings underscore the potential of leveraging organismal growth temperatures in context-dependent design of proteins for enhanced thermal stability.
more »
« less
- Award ID(s):
- 2320718
- PAR ID:
- 10584481
- Publisher / Repository:
- Nature Publishing Group
- Date Published:
- Journal Name:
- Scientific Reports
- Volume:
- 15
- Issue:
- 1
- ISSN:
- 2045-2322
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
SUMMARY Identification of protein interactors is ideally suited for the functional characterization of small molecules. 3′,5′‐cAMP is an evolutionary ancient signaling metabolite largely uncharacterized in plants. To tap into the physiological roles of 3′,5′‐cAMP, we used a chemo‐proteomics approach, thermal proteome profiling (TPP), for the unbiased identification of 3′,5′‐cAMP protein targets. TPP measures shifts in the protein thermal stability upon ligand binding. Comprehensive proteomics analysis yielded a list of 51 proteins significantly altered in their thermal stability upon incubation with 3′,5′‐cAMP. The list contained metabolic enzymes, ribosomal subunits, translation initiation factors, and proteins associated with the regulation of plant growth such as CELL DIVISION CYCLE 48. To functionally validate obtained results, we focused on the role of 3′,5′‐cAMP in regulating the actin cytoskeleton suggested by the presence of actin among the 51 identified proteins. 3′,5′‐cAMP supplementation affected actin organization by inducing actin‐bundling. Consistent with these results, the increase in 3′,5′‐cAMP levels, obtained either by feeding or by chemical modulation of 3′,5′‐cAMP metabolism, was sufficient to partially rescue the short hypocotyl phenotype of theactin2 actin7mutant, severely compromised in actin level. The observed rescue was specific to 3′,5′‐cAMP, as demonstrated using a positional isomer 2′,3′‐cAMP, and true for the nanomolar 3′,5′‐cAMP concentrations reported for plant cells.In vitrocharacterization of the 3′,5′‐cAMP–actin pairing argues against a direct interaction between actin and 3′,5′‐cAMP. Alternative mechanisms by which 3′,5′‐cAMP would affect actin dynamics, such as by interfering with calcium signaling, are discussed. In summary, our work provides a specific resource, 3′,5′‐cAMP interactome, as well as functional insight into 3′,5′‐cAMP‐mediated regulation in plants.more » « less
-
We examine changes in the picosecond structural dynamics with irreversible photobleaching of red fluorescent proteins (RFP) mCherry, mOrange2 and TagRFP-T. Measurements of the protein dynamical transition using terahertz time-domain spectroscopy show in all cases an increase in the turn-on temperature in the bleached state. The result is surprising given that there is little change in the protein surface, and thus, the solvent dynamics held responsible for the transition should not change. A spectral analysis of the measurements guided by quasiharmonic calculations of the protein absorbance reveals that indeed the solvent dynamical turn-on temperature is independent of the thermal stability/photostate however the protein dynamical turn-on temperature shifts to higher temperatures. This is the first demonstration of switching the protein dynamical turn-on temperature with protein functional state. The observed shift in protein dynamical turn-on temperature relative to the solvent indicates an increase in the required mobile waters necessary for the protein picosecond motions, that is, these motions are more collective. Melting-point measurements reveal that the photobleached state is more thermally stable, and structural analysis of related RFP’s shows that there is an increase in internal water channels as well as a more uniform atomic root mean squared displacement. These observations are consistent with previous suggestions that water channels form with extended light excitation providing O2 access to the chromophore and subsequent fluorescence loss. We report that these same channels increase internal coupling enhancing thermal stability and collectivity of the picosecond protein motions. The terahertz spectroscopic characterization of the protein and solvent dynamical onsets can be applied generally to measure changes in collectivity of protein motions.more » « less
-
Abstract Global warming is one of the most significant and widespread effects of climate change. While early life stages are particularly vulnerable to increasing temperatures, little is known about the molecular processes that underpin their capacity to adapt to temperature change during early development. Using a quantitative proteomics approach, we investigated the effects of thermal stress on octopus embryos. We exposedOctopus berrimaembryos to different temperature treatments (control 19°C, current summer temperature 22°C, or future projected summer temperature 25°C) until hatching. By comparing their protein expression levels, we found that future projected temperatures significantly reduced levels of key eye proteins such as S‐crystallin and retinol dehydrogenase 12, suggesting the embryonic octopuses had impaired vision at elevated temperature. We also found that this was coupled with a cellular stress response that included a significant elevation of proteins involved in molecular chaperoning and redox regulation. Energy resources were also redirected away from non‐essential processes such as growth and digestion. These findings, taken together with the high embryonic mortality observed under the highest temperature, identify critical physiological functions of embryonic octopuses that may be impaired under future warming conditions. Our findings demonstrate the severity of the thermal impacts on the early life stages of octopuses as demonstrated by quantitative proteome changes that affect vision, protein chaperoning, redox regulation and energy metabolism as critical physiological functions that underlie the responses to thermal stress.more » « less
-
Abstract Understanding protein function and developing molecular therapies require deciphering the cell types in which proteins act as well as the interactions between proteins. However, modeling protein interactions across biological contexts remains challenging for existing algorithms. Here we introduce PINNACLE, a geometric deep learning approach that generates context-aware protein representations. Leveraging a multiorgan single-cell atlas,PINNACLElearns on contextualized protein interaction networks to produce 394,760 protein representations from 156 cell type contexts across 24 tissues.PINNACLE’s embedding space reflects cellular and tissue organization, enabling zero-shot retrieval of the tissue hierarchy. Pretrained protein representations can be adapted for downstream tasks: enhancing 3D structure-based representations for resolving immuno-oncological protein interactions, and investigating drugs’ effects across cell types.PINNACLEoutperforms state-of-the-art models in nominating therapeutic targets for rheumatoid arthritis and inflammatory bowel diseases and pinpoints cell type contexts with higher predictive capability than context-free models.PINNACLE’s ability to adjust its outputs on the basis of the context in which it operates paves the way for large-scale context-specific predictions in biology.more » « less
