Fueling ab initio folding with marine metagenomics enables structure and function predictions of new protein families

Petabase-Scale Homology Search for Structure Prediction

The recent CASP15 competition highlighted the critical role of multiple sequence alignments (MSAs) in protein structure prediction, as demonstrated by the success of the top AlphaFold2-based prediction methods. To push the boundaries of MSA utilization, we conducted a petabase-scale search of the Sequence Read Archive (SRA), resulting in gigabytes of aligned homologs for CASP15 targets. These were merged with default MSAs produced by ColabFold-search and provided to ColabFold-predict. By using SRA data, we achieved highly accurate predictions (GDT_TS > 70) for 66% of the non-easy targets, whereas using ColabFold-search default MSAs scored highly in only 52%. Next, we tested the effect of deep homology search and ColabFold's advanced features, such as more recycles, on prediction accuracy. While SRA homologs were most significant for improving ColabFold's CASP15 ranking from 11th to 3rd place, other strategies contributed too. We analyze these in the context of existing strategies to improve prediction.

Improving deep learning protein monomer and complex structure prediction using DeepMSA2 with huge metagenomics data

Leveraging iterative alignment search through genomic and metagenome sequence databases, we report the DeepMSA2 pipeline for uniform protein single- and multichain multiple-sequence alignment (MSA) construction. Large-scale benchmarks show that DeepMSA2 MSAs can remarkably increase the accuracy of protein tertiary and quaternary structure predictions compared with current state-of-the-art methods. An integrated pipeline with DeepMSA2 participated in the most recent CASP15 experiment and created complex structural models with considerably higher quality than the AlphaFold2-Multimer server (v.2.2.0). Detailed data analyses show that the major advantage of DeepMSA2 lies in its balanced alignment search and effective model selection, and in the power of integrating huge metagenomics databases. These results demonstrate a new avenue to improve deep learning protein structure prediction through advanced MSA construction and provide additional evidence that optimization of input information to deep learning-based structure prediction methods must be considered with as much care as the design of the predictor itself.

The origin and structural evolution of de novo genes in Drosophila

Recent studies reveal that de novo gene origination from previously non-genic sequences is a common mechanism for gene innovation. These young genes provide an opportunity to study the structural and functional origins of proteins. Here, we combine high-quality base-level whole-genome alignments and computational structural modeling to study the origination, evolution, and protein structures of lineage-specific de novo genes. We identify 555 de novo gene candidates in D. melanogaster that originated within the Drosophilinae lineage. Sequence composition, evolutionary rates, and expression patterns indicate possible gradual functional or adaptive shifts with their gene ages. Surprisingly, we find little overall protein structural changes in candidates from the Drosophilinae lineage. We identify several candidates with potentially well-folded protein structures. Ancestral sequence reconstruction analysis reveals that most potentially well-folded candidates are often born well-folded. Single-cell RNA-seq analysis in testis shows that although most de novo gene candidates are enriched in spermatocytes, several young candidates are biased towards the early spermatogenesis stage, indicating potentially important but less emphasized roles of early germline cells in the de novo gene origination in testis. This study provides a systematic overview of the origin, evolution, and protein structural changes of Drosophilinae-specific de novo genes.

Precision enzyme discovery through targeted mining of metagenomic data

Metagenomics has opened new avenues for exploring the genetic potential of uncultured microorganisms, which may serve as promising sources of enzymes and natural products for industrial applications. Identifying enzymes with improved catalytic properties from the vast amount of available metagenomic data poses a significant challenge that demands the development of novel computational and functional screening tools. The catalytic properties of all enzymes are primarily dictated by their structures, which are predominantly determined by their amino acid sequences. However, this aspect has not been fully considered in the enzyme bioprospecting processes. With the accumulating number of available enzyme sequences and the increasing demand for discovering novel biocatalysts, structural and functional modeling can be employed to identify potential enzymes with novel catalytic properties. Recent efforts to discover new polysaccharide-degrading enzymes from rumen metagenome data using homology-based searches and machine learning-based models have shown significant promise. Here, we will explore various computational approaches that can be employed to screen and shortlist metagenome-derived enzymes as potential biocatalyst candidates, in conjunction with the wet lab analytical methods traditionally used for enzyme characterization.

Integrating deep learning, threading alignments, and a multi-MSA strategy for high-quality protein monomer and complex structure prediction in CASP15

We report the results of the "UM-TBM" and "Zheng" groups in CASP15 for protein monomer and complex structure prediction. These prediction sets were obtained using the D-I-TASSER and DMFold-Multimer algorithms, respectively. For monomer structure prediction, D-I-TASSER introduced four new features during CASP15: (i) a multiple sequence alignment (MSA) generation protocol that combines multi-source MSA searching and a structural modeling-based MSA ranker; (ii) attention-network based spatial restraints; (iii) a multi-domain module containing domain partition and arrangement for domain-level templates and spatial restraints; (iv) an optimized I-TASSER-based folding simulation system for full-length model creation guided by a combination of deep learning restraints, threading alignments, and knowledge-based potentials. For 47 free modeling targets in CASP15, the final models predicted by D-I-TASSER showed average TM-score 19% higher than the standard AlphaFold2 program. We thus showed that traditional Monte Carlo-based folding simulations, when appropriately coupled with deep learning algorithms, can generate models with improved accuracy over end-to-end deep learning methods alone. For protein complex structure prediction, DMFold-Multimer generated models by integrating a new MSA generation algorithm (DeepMSA2) with the end-to-end modeling module from AlphaFold2-Multimer. For the 38 complex targets, DMFold-Multimer generated models with an average TM-score of 0.83 and Interface Contact Score of 0.60, both significantly higher than those of competing complex prediction tools. Our analyses on complexes highlighted the critical role played by MSA generating, ranking, and pairing in protein complex structure prediction. We also discuss future room for improvement in the areas of viral protein modeling and complex model ranking.

Petascale Homology Search for Structure Prediction

The recent CASP15 competition highlighted the critical role of multiple sequence alignments (MSAs) in protein structure prediction, as demonstrated by the success of the top AlphaFold2-based prediction methods. To push the boundaries of MSA utilization, we conducted a petabase-scale search of the Sequence Read Archive (SRA), resulting in gigabytes of aligned homologs for CASP15 targets. These were merged with default MSAs produced by ColabFold-search and provided to ColabFold-predict. By using SRA data, we achieved highly accurate predictions (GDT_TS > 70) for 66% of the non-easy targets, whereas using ColabFold-search default MSAs scored highly in only 52%. Next, we tested the effect of deep homology search and ColabFold's advanced features, such as more recycles, on prediction accuracy. While SRA homologs were most significant for improving ColabFold's CASP15 ranking from 11th to 3rd place, other strategies contributed too. We analyze these in the context of existing strategies to improve prediction.

Microbiome-based enrichment pattern mining has enabled a deeper understanding of the biome-species-function relationship

Microbes live in diverse habitats (i.e. biomes), yet their species and genes were biome-specific, forming enrichment patterns. These enrichment patterns have mirrored the biome-species-function relationship, which is shaped by ecological and evolutionary principles. However, a grand picture of these enrichment patterns, as well as the roles of external and internal factors in driving these enrichment patterns, remain largely unexamined. In this work, we have examined the enrichment patterns based on 1705 microbiome samples from four representative biomes (Engineered, Gut, Freshwater, and Soil). Moreover, an "enrichment sphere" model was constructed to elucidate the regulatory principles behind these patterns. The driving factors for this model were revealed based on two case studies: (1) The copper-resistance genes were enriched in Soil biomes, owing to the copper contamination and horizontal gene transfer. (2) The flagellum-related genes were enriched in the Freshwater biome, due to high fluidity and vertical gene accumulation. Furthermore, this enrichment sphere model has valuable applications, such as in biome identification for metagenome samples, and in guiding 3D structure modeling of proteins. In summary, the enrichment sphere model aims towards creating a bluebook of the biome-species-function relationships and be applied in many fields.

Contact-Assisted Threading in Low-Homology Protein Modeling

The ability to successfully predict the three-dimensional structure of a protein from its amino acid sequence has made considerable progress in the recent past. The progress is propelled by the improved accuracy of deep learning-based inter-residue contact map predictors coupled with the rising growth of protein sequence databases. Contact map encodes interatomic interaction information that can be exploited for highly accurate prediction of protein structures via contact map threading even for the query proteins that are not amenable to direct homology modeling. As such, contact-assisted threading has garnered considerable research effort. In this chapter, we provide an overview of existing contact-assisted threading methods while highlighting the recent advances and discussing some of the current limitations and future prospects in the application of contact-assisted threading for improving the accuracy of low-homology protein modeling.

Exploring microbial functional biodiversity at the protein family level-From metagenomic sequence reads to annotated protein clusters

Metagenomics has enabled accessing the genetic repertoire of natural microbial communities. Metagenome shotgun sequencing has become the method of choice for studying and classifying microorganisms from various environments. To this end, several methods have been developed to process and analyze the sequence data from raw reads to end-products such as predicted protein sequences or families. In this article, we provide a thorough review to simplify such processes and discuss the alternative methodologies that can be followed in order to explore biodiversity at the protein family level. We provide details for analysis tools and we comment on their scalability as well as their advantages and disadvantages. Finally, we report the available data repositories and recommend various approaches for protein family annotation related to phylogenetic distribution, structure prediction and metadata enrichment.

Ontology-aware deep learning enables ultrafast and interpretable source tracking among sub-million microbial community samples from hundreds of niches

The taxonomic structure of microbial community sample is highly habitat-specific, making source tracking possible, allowing identification of the niches where samples originate. However, current methods face challenges when source tracking is scaled up. Here, we introduce a deep learning method based on the Ontology-aware Neural Network approach, ONN4MST, for large-scale source tracking. ONN4MST outperformed other methods with near-optimal accuracy when source tracking among 125,823 samples from 114 niches. ONN4MST also has a broad spectrum of applications. Overall, this study represents the first model-based method for source tracking among sub-million microbial community samples from hundreds of niches, with superior speed, accuracy, and interpretability. ONN4MST is available at https://github.com/HUST-NingKang-Lab/ONN4MST.

MrParse: finding homologues in the PDB and the EBI AlphaFold database for molecular replacement and more

Crystallographers have an array of search-model options for structure solution by molecular replacement (MR). The well established options of homologous experimental structures and regular secondary-structure elements or motifs are increasingly supplemented by computational modelling. Such modelling may be carried out locally or may use pre-calculated predictions retrieved from databases such as the EBI AlphaFold database. MrParse is a new pipeline to help to streamline the decision process in MR by consolidating bioinformatic predictions in one place. When reflection data are provided, MrParse can rank any experimental homologues found using eLLG, which indicates the likelihood that a given search model will work in MR. Inbuilt displays of predicted secondary structure, coiled-coil and transmembrane regions further inform the choice of MR protocol. MrParse can also identify and rank homologues in the EBI AlphaFold database, a function that will also interest other structural biologists and bioinformaticians.

Structural insight into protein-protein interactions between intestinal microbiome and host

Protein-protein interactions between the microbiome and host organism play an important role in shaping host health. These host-modulating proteins have therapeutic potential in treating microbiome-linked disorders such as inflammatory bowel disease and obesity. Structural analysis of interacting proteins provides highly mechanistic insight into the domains driving these interactions and the resulting influence on host cell processes. Here, we briefly review recent publication of microbiome protein structures involved in host binding interactions, the effects of these interactions on host physiology, and the need for further study to increase the ability to detect proteins with therapeutic potential.

How much metagenome data is needed for protein structure prediction: The advantages of targeted approach from the ecological and evolutionary perspectives

It has been proven that three-dimensional protein structures could be modeled by supplementing homologous sequences with metagenome sequences. Even though a large volume of metagenome data is utilized for such purposes, a significant proportion of proteins remain unsolved. In this review, we focus on identifying ecological and evolutionary patterns in metagenome data, decoding the complicated relationships of these patterns with protein structures, and investigating how these patterns can be effectively used to improve protein structure prediction. First, we proposed the metagenome utilization efficiency and marginal effect model to quantify the divergent distribution of homologous sequences for the protein family. Second, we proposed that the targeted approach effectively identifies homologous sequences from specified biomes compared with the untargeted approach's blind search. Finally, we determined the lower bound for metagenome data required for predicting all the protein structures in the Pfam database and showed that the present metagenome data is insufficient for this purpose. In summary, we discovered ecological and evolutionary patterns in the metagenome data that may be used to predict protein structures effectively. The targeted approach is promising in terms of effectively extracting homologous sequences and predicting protein structures using these patterns.

Using metagenomic data to boost protein structure prediction and discovery

Over the past decade, metagenomic sequencing approaches have been providing an ever-increasing amount of protein sequence data at an astonishing rate. These constitute an invaluable source of information which has been exploited in various research fields such as the study of the role of the gut microbiota in human diseases and aging. However, only a small fraction of all metagenomic sequences collected have been functionally or structurally characterized, leaving much of them completely unexplored. Here, we review how this information has been used in protein structure prediction and protein discovery. We begin by presenting some widely used metagenomic databases and analyze in detail how metagenomic data has contributed to the impressive improvement in the accuracy of structure prediction methods in recent years. We then examine how metagenomic information can be exploited to annotate protein sequences. More specifically, we focus on the role of metagenomes in the discovery of enzymes and new CRISPR-Cas systems, and in the identification of antibiotic resistance genes. With this review, we provide an overview of how metagenomic data is currently revolutionizing our understanding of protein science.

Decoding the link of microbiome niches with homologous sequences enables accurately targeted protein structure prediction

Information derived from metagenome sequences through deep-learning techniques has significantly improved the accuracy of template free protein structure modeling. However, most of the deep learning-based modeling studies are based on blind sequence database searches and suffer from low efficiency in computational resource utilization and model construction, especially when the sequence library becomes prohibitively large. We proposed a MetaSource model built on 4.25 billion microbiome sequences from four major biomes (Gut, Lake, Soil, and Fermentor) to decode the inherent linkage of microbial niches with protein homologous families. Large-scale protein family folding experiments on 8,700 unknown Pfam families showed that a microbiome targeted approach with multiple sequence alignment constructed from individual MetaSource biomes requires more than threefold less computer memory and CPU (central processing unit) time but generates contact-map and three-dimensional structure models with a significantly higher accuracy, compared with that using combined metagenome datasets. These results demonstrate an avenue to bridge the gap between the rapidly increasing metagenome databases and the limited computing resources for efficient genome-wide database mining, which provides a useful bluebook to guide future microbiome sequence database and modeling development for high-accuracy protein structure and function prediction.

A roadmap for metagenomic enzyme discovery

Covering: up to 2021Metagenomics has yielded massive amounts of sequencing data offering a glimpse into the biosynthetic potential of the uncultivated microbial majority. While genome-resolved information about microbial communities from nearly every environment on earth is now available, the ability to accurately predict biocatalytic functions directly from sequencing data remains challenging. Compared to primary metabolic pathways, enzymes involved in secondary metabolism often catalyze specialized reactions with diverse substrates, making these pathways rich resources for the discovery of new enzymology. To date, functional insights gained from studies on environmental DNA (eDNA) have largely relied on PCR- or activity-based screening of eDNA fragments cloned in fosmid or cosmid libraries. As an alternative, shotgun metagenomics holds underexplored potential for the discovery of new enzymes directly from eDNA by avoiding common biases introduced through PCR- or activity-guided functional metagenomics workflows. However, inferring new enzyme functions directly from eDNA is similar to searching for a 'needle in a haystack' without direct links between genotype and phenotype. The goal of this review is to provide a roadmap to navigate shotgun metagenomic sequencing data and identify new candidate biosynthetic enzymes. We cover both computational and experimental strategies to mine metagenomes and explore protein sequence space with a spotlight on natural product biosynthesis. Specifically, we compare in silico methods for enzyme discovery including phylogenetics, sequence similarity networks, genomic context, 3D structure-based approaches, and machine learning techniques. We also discuss various experimental strategies to test computational predictions including heterologous expression and screening. Finally, we provide an outlook for future directions in the field with an emphasis on meta-omics, single-cell genomics, cell-free expression systems, and sequence-independent methods.

Toward the solution of the protein structure prediction problem

Since Anfinsen demonstrated that the information encoded in a protein's amino acid sequence determines its structure in 1973, solving the protein structure prediction problem has been the Holy Grail of structural biology. The goal of protein structure prediction approaches is to utilize computational modeling to determine the spatial location of every atom in a protein molecule starting from only its amino acid sequence. Depending on whether homologous structures can be found in the Protein Data Bank (PDB), structure prediction methods have been historically categorized as template-based modeling (TBM) or template-free modeling (FM) approaches. Until recently, TBM has been the most reliable approach to predicting protein structures, and in the absence of reliable templates, the modeling accuracy sharply declines. Nevertheless, the results of the most recent community-wide assessment of protein structure prediction experiment (CASP14) have demonstrated that the protein structure prediction problem can be largely solved through the use of end-to-end deep machine learning techniques, where correct folds could be built for nearly all single-domain proteins without using the PDB templates. Critically, the model quality exhibited little correlation with the quality of available template structures, as well as the number of sequence homologs detected for a given target protein. Thus, the implementation of deep-learning techniques has essentially broken through the 50-year-old modeling border between TBM and FM approaches and has made the success of high-resolution structure prediction significantly less dependent on template availability in the PDB library.

Recent Advances in Protein Homology Detection Propelled by Inter-Residue Interaction Map Threading

Sequence-based protein homology detection has emerged as one of the most sensitive and accurate approaches to protein structure prediction. Despite the success, homology detection remains very challenging for weakly homologous proteins with divergent evolutionary profile. Very recently, deep neural network architectures have shown promising progress in mining the coevolutionary signal encoded in multiple sequence alignments, leading to reasonably accurate estimation of inter-residue interaction maps, which serve as a rich source of additional information for improved homology detection. Here, we summarize the latest developments in protein homology detection driven by inter-residue interaction map threading. We highlight the emerging trends in distant-homology protein threading through the alignment of predicted interaction maps at various granularities ranging from binary contact maps to finer-grained distance and orientation maps as well as their combination. We also discuss some of the current limitations and possible future avenues to further enhance the sensitivity of protein homology detection.

Microbiome Sample Comparison and Search: From Pair-Wise Calculations to Model-Based Matching

A huge quantity of microbiome samples have been accumulated, and more are yet to come from all niches around the globe. With the accumulation of data, there is an urgent need for comparisons and searches of microbiome samples among thousands of millions of samples in a fast and accurate manner. However, it is a very difficult computational challenge to identify similar samples, as well as identify their likely origins, among such a grand pool of samples from all around the world. Currently, several approaches have already been proposed for such a challenge, based on either distance calculation, unsupervised algorithms, or supervised algorithms. These methods have advantages and disadvantages for the different settings of comparisons and searches, and their results are also drastically different. In this review, we systematically compared distance-based, unsupervised, and supervised methods for microbiome sample comparison and search. Firstly, we assessed their accuracy and efficiency, both in theory and in practice. Then we described the scenarios in which one or multiple methods were applicable for sample searches. Thirdly, we provided several applications for microbiome sample comparisons and searches, and provided suggestions on the choice of methods. Finally, we provided several perspectives for the future development of microbiome sample comparison and search, including deep learning technologies for tracking the sources of microbiome samples.

DeepMSA: constructing deep multiple sequence alignment to improve contact prediction and fold-recognition for distant-homology proteins

MOTIVATION

The success of genome sequencing techniques has resulted in rapid explosion of protein sequences. Collections of multiple homologous sequences can provide critical information to the modeling of structure and function of unknown proteins. There are however no standard and efficient pipeline available for sensitive multiple sequence alignment (MSA) collection. This is particularly challenging when large whole-genome and metagenome databases are involved.

RESULTS

We developed DeepMSA, a new open-source method for sensitive MSA construction, which has homologous sequences and alignments created from multi-sources of whole-genome and metagenome databases through complementary hidden Markov model algorithms. The practical usefulness of the pipeline was examined in three large-scale benchmark experiments based on 614 non-redundant proteins. First, DeepMSA was utilized to generate MSAs for residue-level contact prediction by six coevolution and deep learning-based programs, which resulted in an accuracy increase in long-range contacts by up to 24.4% compared to the default programs. Next, multiple threading programs are performed for homologous structure identification, where the average TM-score of the template alignments has over 7.5% increases with the use of the new DeepMSA profiles. Finally, DeepMSA was used for secondary structure prediction and resulted in statistically significant improvements in the Q3 accuracy. It is noted that all these improvements were achieved without re-training the parameters and neural-network models, demonstrating the robustness and general usefulness of the DeepMSA in protein structural bioinformatics applications, especially for targets without homologous templates in the PDB library.

AVAILABILITY AND IMPLEMENTATION

https://zhanglab.ccmb.med.umich.edu/DeepMSA/.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

State-of-the-art web services for de novo protein structure prediction

Residue coevolution estimations coupled to machine learning methods are revolutionizing the ability of protein structure prediction approaches to model proteins that lack clear homologous templates in the Protein Data Bank (PDB). This has been patent in the last round of the Critical Assessment of Structure Prediction (CASP), which presented several very good models for the hardest targets. Unfortunately, literature reporting on these advances often lacks digests tailored to lay end users; moreover, some of the top-ranking predictors do not provide webservers that can be used by nonexperts. How can then end users benefit from these advances and correctly interpret the predicted models? Here we review the web resources that biologists can use today to take advantage of these state-of-the-art methods in their research, including not only the best de novo modeling servers but also datasets of models precomputed by experts for structurally uncharacterized protein families. We highlight their features, advantages and pitfalls for predicting structures of proteins without clear templates. We present a broad number of applications that span from driving forward biochemical investigations that lack experimental structures to actually assisting experimental structure determination in X-ray diffraction, cryo-EM and other forms of integrative modeling. We also discuss issues that must be considered by users yet still require further developments, such as global and residue-wise model quality estimates and sources of residue coevolution other than monomeric tertiary structure.

Fueling ab initio folding with marine metagenomics enables structure and function predictions of new protein families

INTRODUCTION

The ocean microbiome represents one of the largest microbiomes and produces nearly half of the primary energy on the planet through photosynthesis or chemosynthesis. Using recent advances in marine genomics, we explore new applications of oceanic metagenomes for protein structure and function prediction.

RESULTS

By processing 1.3 TB of high-quality reads from the Tara Oceans data, we obtain 97 million non-redundant genes. Of the 5721 Pfam families that lack experimental structures, 2801 have at least one member associated with the oceanic metagenomics dataset. We apply C-QUARK, a deep-learning contact-guided ab initio structure prediction pipeline, to model 27 families, where 20 are predicted to have a reliable fold with estimated template modeling score (TM-score) at least 0.5. Detailed analyses reveal that the abundance of microbial genera in the ocean is highly correlated to the frequency of occurrence in the modeled Pfam families, suggesting the significant role of the Tara Oceans genomes in the contact-map prediction and subsequent ab initio folding simulations. Of interesting note, PF15461, which has a majority of members coming from ocean-related bacteria, is identified as an important photosynthetic protein by structure-based function annotations. The pipeline is extended to a set of 417 Pfam families, built on the combination of Tara with other metagenomics datasets, which results in 235 families with an estimated TM-score over 0.5.

CONCLUSIONS

These results demonstrate a new avenue to improve the capacity of protein structure and function modeling through marine metagenomics, especially for difficult proteins with few homologous sequences.