A labeled dataset for AI-based cryo-EM map enhancement
Not provided.
Biochemistry & Molecular Biology↗
Engineering topics
Cheng, Jianlin
Publications and source records attributed to Cheng, Jianlin.
De novo atomic protein structure modeling for cryoEM density maps using 3D transformer and HMM
Accurately building 3D atomic structures from cryo-EM density maps is a crucial step in cryo-EM-based protein structure determination. Converting density maps into 3D atomic structures for proteins lacking accurate homologous or predicted structures as templates remains a significant challenge. Here, we introduce Cryo2Struct, a fully automated de novo cryo-EM structure modeling method. Cryo2Struct utilizes a 3D transformer to identify atoms and amino acid types in cryo-EM density maps, followed by an innovative Hidden Markov Model (HMM) to connect predicted atoms and build protein backbone structures. Cryo2Struct produces substantially more accurate and complete protein structural models than the widely used ab initio method Phenix. Additionally, its performance in building atomic structural models is robust against changes in the resolution of density maps and the size of protein structures.
59 BASIC BIOLOGICAL SCIENCES↗
CryoSegNet: accurate cryo-EM protein particle picking by integrating the foundational AI image segmentation model and attention-gated U-Net
Picking protein particles in cryo-electron microscopy (cryo-EM) micrographs is a crucial step in the cryo-EM-based structure determination. However, existing methods trained on a limited amount of cryo-EM data still cannot accurately pick protein particles from noisy cryo-EM images. The general foundational artificial intelligence–based image segmentation model such as Meta’s Segment Anything Model (SAM) cannot segment protein particles well because their training data do not include cryo-EM images. Here, we present a novel approach (CryoSegNet) of integrating an attention-gated U-shape network (U-Net) specially designed and trained for cryo-EM particle picking and the SAM. The U-Net is first trained on a large cryo-EM image dataset and then used to generate input from original cryo-EM images for SAM to make particle pickings. CryoSegNet shows both high precision and recall in segmenting protein particles from cryo-EM micrographs, irrespective of protein type, shape and size. On several independent datasets of various protein types, CryoSegNet outperforms two top machine learning particle pickers crYOLO and Topaz as well as SAM itself. The average resolution of density maps reconstructed from the particles picked by CryoSegNet is 3.33 Å, 7% better than 3.58 Å of Topaz and 14% better than 3.87 Å of crYOLO. It is publicly available at https://github.com/jianlin-cheng/CryoSegNet
59 BASIC BIOLOGICAL SCIENCES↗
High throughput, accurate gene annotation through AI and HPC-enabled structural analysis
With the advances in next generation sequencing technologies, the number of sequenced genomes is growing exponentially, resulting in a technology bottleneck for the translation of sequence information into usable hypotheses about the function of each gene. We have proposed leveraging our leadership high-performance computing (HPC) resources to help break this annotation bottleneck. Here we design an HPC-based framework to infer gene function from gene sequence by incorporating information about protein structure and interactions predicted by deep learning approaches. Accurate functional prediction and gene annotation using computational methods will facilitate breakthroughs in the genomic sciences essential to understanding and harnessing life processes in bacteria, fungi and plants. The development and applications of the state-of-the-art deep neural networks to protein structural modeling, interaction prediction, sequence comparison, and quality assessment of protein structural models will be made possible by leadership computational resources. These HPC-enabled bioinformatics and molecular modeling tools will provide powerful insights into molecular functions of genes.
59 BASIC BIOLOGICAL SCIENCES↗
Deep Learning Prediction of Protein Complex Structures
Proteins interact to form protein complex to carry out biological functions such as catalytic chemical reaction. Therefore, it is important to develop computational methods to predict protein-protein interaction and the structures of protein complexes to study and enhance protein function. In this project, we successfully developed several deep learning methods to predict inter-protein contacts and the reinforcement learning and optimization methods to reconstruct protein complex structures from predicted inter-chain contacts. The methods were integrated with the MULTICOM protein complex structure prediction system and applied to predict the complex structures of biomass production-related proteins of green algae. During the two and a half years of research and development, all the specific milestones of the project were achieved successfully. 16 publications/manuscripts were produced. 10 software tools were developed. A patent application was submitted. Our MULTICOM predictors leveraging some tools developed in this project were ranked among the top predictors in the 15th Critical Assessment of Techniques for Protein Structure Prediction (CASP15) in 2022.
59 BASIC BIOLOGICAL SCIENCES↗
Cryo2StructData: A Large Labeled Cryo-EM Density Map Dataset for AI-based Modeling of Protein Structures
The advent of single-particle cryo-electron microscopy (cryo-EM) has brought forth a new era of structural biology, enabling the routine determination of large biological molecules and their complexes at atomic resolution. The high-resolution structures of biological macromolecules and their complexes significantly expedite biomedical research and drug discovery. However, automatically and accurately building atomic models from high-resolution cryo-EM density maps is still time-consuming and challenging when template-based models are unavailable. Artificial intelligence (AI) methods such as deep learning trained on limited amount of labeled cryo-EM density maps generate inaccurate atomic models. To address this issue, we created a dataset called Cryo2StructData consisting of 7,600 preprocessed cryo-EM density maps whose voxels are labelled according to their corresponding known atomic structures for training and testing AI methods to build atomic models from cryo-EM density maps. Cryo2StructData is larger than existing, publicly available datasets for training AI methods to build atomic protein structures from cryo-EM density maps. We trained and tested deep learning models on Cryo2StructData to validate its quality showing that it is ready for being used to train and test AI methods for building atomic models.
59 BASIC BIOLOGICAL SCIENCES↗
Biochemical, structural, and computational analyses of two new clinically identified missense mutations of ALDH7A1
Not provided.
Biochemistry & Molecular Biology↗
CryoTransformer: a transformer model for picking protein particles from Cryo-EM micrographs
Cryo-electron microscopy (cryo-EM) is a powerful technique for determining the structures of large protein complexes. Picking single protein particles from cryo-EM micrographs (images) is a crucial step in reconstructing protein structures from them. However, the widely used template-based particle picking process requires some manual particle picking and is labor-intensive and time-consuming. Though machine learning and artificial intelligence (AI) can potentially automate particle picking, the current AI methods pick particles with low precision or low recall. The erroneously picked particles can severely reduce the quality of reconstructed protein structures, especially for the micrographs with low signal-to-noise-ratio (SNR).
59 BASIC BIOLOGICAL SCIENCES↗
Identification and preliminary characterization of conserved uncharacterized proteins from Chlamydomonas reinhardtii , Arabidopsis thaliana , and Setaria viridis
Abstract The rapid accumulation of sequenced plant genomes in the past decade has outpaced the still difficult problem of genome‐wide protein‐coding gene annotation. A substantial fraction of protein‐coding genes in all plant genomes are poorly annotated or unannotated and remain functionally uncharacterized. We identified unannotated proteins in three model organisms representing distinct branches of the green lineage (Viridiplantae): Arabidopsis thaliana (eudicot), Setaria viridis (monocot), and Chlamydomonas reinhardtii (Chlorophyte alga). Using similarity searching, we identified a subset of unannotated proteins that were conserved between these species and defined them as Deep Green proteins. Bioinformatic, genomic, and structural predictions were performed to begin classifying Deep Green genes and proteins. Compared to whole proteomes for each species, the Deep Green set was enriched for proteins with predicted chloroplast targeting signals predictive of photosynthetic or plastid functions, a result that was consistent with enrichment for daylight phase diurnal expression patterning. Structural predictions using AlphaFold and comparisons to known structures showed that a significant proportion of Deep Green proteins may possess novel folds. Though only available for three organisms, the Deep Green genes and proteins provide a starting resource of high‐value targets for further investigation of potentially new protein structures and functions conserved across the green lineage.
59 BASIC BIOLOGICAL SCIENCES↗
Enhancing alphafold-multimer-based protein complex structure prediction with MULTICOM in CASP15
To enhance the AlphaFold-Multimer-based protein complex structure prediction, we developed a quaternary structure prediction system (MULTICOM) to improve the input fed to AlphaFold-Multimer and evaluate and refine its outputs. MULTICOM samples diverse multiple sequence alignments (MSAs) and templates for AlphaFold-Multimer to generate structural predictions by using both traditional sequence alignments and Foldseek-based structure alignments, ranks structural predictions through multiple complementary metrics, and refines the structural predictions via a Foldseek structure alignment-based refinement method. The MULTICOM system with different implementations was blindly tested in the assembly structure prediction in the 15th Critical Assessment of Techniques for Protein Structure Prediction (CASP15) in 2022 as both server and human predictors. MULTICOM_qa ranked 3 rd among 26 CASP15 server predictors and MULTICOM_human ranked 7 th among 87 CASP15 server and human predictors. The average TM-score of the first predictions submitted by MULTICOM_qa for CASP15 assembly targets is ~0.76, 5.3% higher than ~0.72 of the standard AlphaFold-Multimer. The average TM-score of the best of top 5 predictions submitted by MULTICOM_qa is ~0.80, about 8% higher than ~0.74 of the standard AlphaFold-Multimer. Moreover, the Foldseek Structure Alignment-based Multimer structure Generation (FSAMG) method outperforms the widely used sequence alignment-based multimer structure generation.
59 BASIC BIOLOGICAL SCIENCES↗
Improving AlphaFold2-based protein tertiary structure prediction with MULTICOM in CASP15
Since the 14th Critical Assessment of Techniques for Protein Structure Prediction (CASP14), AlphaFold2 has become the standard method for protein tertiary structure prediction. One remaining challenge is to further improve its prediction. We developed a new version of the MULTICOM system to sample diverse multiple sequence alignments (MSAs) and structural templates to improve the input for AlphaFold2 to generate structural models. The models are then ranked by both the pairwise model similarity and AlphaFold2 self-reported model quality score. The top ranked models are refined by a novel structure alignment-based refinement method powered by Foldseek. Moreover, for a monomer target that is a subunit of a protein assembly (complex), MULTICOM integrates tertiary and quaternary structure predictions to account for tertiary structural changes induced by protein-protein interaction. The system participated in the tertiary structure prediction in 2022 CASP15 experiment. Our server predictor MULTICOM_refine ranked 3rd among 47 CASP15 server predictors and our human predictor MULTICOM ranked 7th among all 132 human and server predictors. The average GDT-TS score and TM-score of the first structural models that MULTICOM_refine predicted for 94 CASP15 domains are ~0.80 and ~0.92, 9.6% and 8.2% higher than ~0.73 and 0.85 of the standard AlphaFold2 predictor respectively.
59 BASIC BIOLOGICAL SCIENCES↗
Crystal structure of domain of unknown function 507 (DUF507) reveals a new protein fold
The crystal structure of the domain of unknown function family 507 protein from Aquifex aeolicus is reported (AaDUF507, UniProt O67633, 183 residues). The structure was determined in two space groups (C222 1 and P3 2 21) at 1.9 Å resolution. The phase problem was solved by molecular replacement using an AlphaFold model as the search model. AaDUF507 is a Y-shaped α-helical protein consisting of an anti-parallel 4-helix bundle base and two helical arms that extend 30-Å from the base. The two crystal structures differ by a 25° rigid body rotation of the C-terminal arm. The tertiary structure exhibits pseudo-twofold symmetry. The structural symmetry mirrors internal sequence similarity: residues 11–57 and 102–148 are 30% identical and 53% similar with an E-value of 0.002. In one of the structures, electron density for an unknown ligand, consistent with nicotinamide or similar molecule, may indicate a functional site. Docking calculations suggest potential ligand binding hot spots in the region between the helical arms. Structure-based query of the Protein Data Bank revealed no other protein with a similar tertiary structure, leading us to propose that AaDUF507 represents a new protein fold.
36 MATERIALS SCIENCE↗
Predicted structural proteome of Sphagnum divinum and proteome-scale annotation
Sphagnum-dominated peatlands store a substantial amount of terrestrial carbon. The genus is undersampled and under-studied. No experimental crystal structure from any Sphagnum species exists in the Protein Data Bank and fewer than 200 Sphagnum-related genes have structural models available in the AlphaFold Protein Structure Database. Tools and resources are needed to help bridge these gaps, and to enable the analysis of other structural proteomes now made possible by accurate structure prediction. We present the predicted structural proteome (25,134 primary transcripts) of Sphagnum divinum computed using AlphaFold, structural alignment results of all high-confidence models against an annotated nonredundant crystallographic database of over 90,000 structures, a structure-based classification of putative Enzyme Commission (EC) numbers across this proteome, and the computational method to perform this proteome-scale structure-based annotation.
59 BASIC BIOLOGICAL SCIENCES↗
Deep Green: Structural and Functional Genomic Characterization of Conserved Unannotated Green Lineage Proteins
Our overall objective is to improve and increase functional and structural predictions for a growing number of plant proteins of unknown structure and function (the Deep Green proteins), and to make our predictive data useful and accessible to the larger research community. The project is divided into five major objectives which include 1) Assembly and curation of Deep Green candidate protein sets; 2) in silico structural and functional predictions and network analyses; 3) assembly and validation of reverse genetic resources in Chlamydomonas reinhardtii (Chlamydomonas); 4) high throughput functional genomics characterization and prioritization in Chlamydomonas; and 5) structural validation of selected candidates and functional validation in two important reference plant species, Arabidopsis thaliana (Arabidopsis) and Setaria viridis (Setaria).
59 BASIC BIOLOGICAL SCIENCES↗
Combining pairwise structural similarity and deep learning interface contact prediction to estimate protein complex model accuracy in CASP15
Abstract Estimating the accuracy of quaternary structural models of protein complexes and assemblies (EMA) is important for predicting quaternary structures and applying them to studying protein function and interaction. The pairwise similarity between structural models is proven useful for estimating the quality of protein tertiary structural models, but it has been rarely applied to predicting the quality of quaternary structural models. Moreover, the pairwise similarity approach often fails when many structural models are of low quality and similar to each other. To address the gap, we developed a hybrid method (MULTICOM_qa) combining a pairwise similarity score (PSS) and an interface contact probability score (ICPS) based on the deep learning inter‐chain contact prediction for estimating protein complex model accuracy. It blindly participated in the 15th Critical Assessment of Techniques for Protein Structure Prediction (CASP15) in 2022 and performed very well in estimating the global structure accuracy of assembly models. The average per‐target correlation coefficient between the model quality scores predicted by MULTICOM_qa and the true quality scores of the models of CASP15 assembly targets is 0.66. The average per‐target ranking loss in using the predicted quality scores to rank the models is 0.14. It was able to select good models for most targets. Moreover, several key factors (i.e., target difficulty, model sampling difficulty, skewness of model quality, and similarity between good/bad models) for EMA are identified and analyzed. The results demonstrate that combining the multi‐model method (PSS) with the complementary single‐model method (ICPS) is a promising approach to EMA.
59 BASIC BIOLOGICAL SCIENCES↗
A large expert-curated cryo-EM image dataset for machine learning protein particle picking
Cryo-electron microscopy (cryo-EM) is a powerful technique for determining the structures of biological macromolecular complexes. Picking single-protein particles from cryo-EM micrographs is a crucial step in reconstructing protein structures. However, the widely used template-based particle picking process is labor-intensive and time-consuming. Though machine learning and artificial intelligence (AI) based particle picking can potentially automate the process, its development is hindered by lack of large, high-quality labelled training data. To address this bottleneck, we present CryoPPP, a large, diverse, expert-curated cryo-EM image dataset for protein particle picking and analysis. It consists of labelled cryo-EM micrographs (images) of 34 representative protein datasets selected from the Electron Microscopy Public Image Archive (EMPIAR). The dataset is 2.6 terabytes and includes 9,893 high-resolution micrographs with labelled protein particle coordinates. The labelling process was rigorously validated through 2D particle class validation and 3D density map validation with the gold standard. The dataset is expected to greatly facilitate the development of both AI and classical methods for automated cryo-EM protein particle picking.
42 ENGINEERING↗
Deep Green Unannotated Protein Structures
The Deep Green list is based on the identification and curation of conserved unannotated proteins in three green lineage (Viridiplantae) model organisms; Arabidopsis thaliana, Chlamydomonas reinhardtii, and Setaria viridis. Preliminary characterization of Deep Green proteins and genes was done using various informatics tools and published data sets and is presented in Knoshaug, Sun, et al., 2023, submitted. The structures of these unannotated proteins were also predicted using AlphaFold (Jumper et al., 2021). The data deposited here are the AlphaFold structural predictions having the highest pLDDT score and thus identified as the best folded structure (ranked_0). These data enable others to do in-depth structural characterizations to aid in functional characterization leading to deeper understanding of plant biology. References: Jumper, J., Evans, R., Pritzel, A., Green, T., Figurnov, M., Ronneberger, O., Tunyasuvunakool, K., Bates, R., Žídek, A., Potapenko, A., Bridgland, A., Meyer, C., Kohl, S. A. A., Ballard, A. J., Cowie, A., Romera-Paredes, B., Nikolov, S., Jain, R., Adler, J., Back, T., Petersen, S., Reiman, D., Clancy, E., Zielinski, M., Steinegger, M., Pacholska, M., Berghammer, T., Bodenstein, S., Silver, D., Vinyals, O., Senior, A. W., Kavukcuoglu, K., Kohli, P. and Hassabis, D. (2021) Highly accurate protein structure prediction with AlphaFold. Nature, 596:583-589. Knoshaug, E. P., Sun, P., Nag, A., Nguyen, H., Mattoon, E. M., Zhang, N., Liu, J., Chen, C., Cheng, J., Zhang, R., St. John, P., and Umen, J. (submitted) Identification and preliminary characterization of conserved uncharacterized proteins from Chlamydomonas reinhardtii, Arabidopsis thaliana, and Setaria viridis.
09 BIOMASS FUELS↗
3D-equivariant graph neural networks for protein model quality assessment
Quality 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. We 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.
59 BASIC BIOLOGICAL SCIENCES↗