Global discriminative learning for higher-accuracy computational gene prediction

Most ab initio gene predictors use a probabilistic sequence model, typically a hidden Markov model, to combine separately trained models of genomic signals and content. By combining separate models of relevant genomic features, such gene predictors can exploit small training sets and incomplete annotations, and can be trained fairly efficiently. However, that type of piecewise training does not optimize prediction accuracy and has difficulty in accounting for statistical dependencies among different parts of the gene model. With genomic information being created at an ever-increasing rate, it is worth investigating alternative approaches in which many different types of genomic evidence, with complex statistical dependencies, can be integrated by discriminative learning to maximize annotation accuracy. Among discriminative learning methods, large-margin classifiers have become prominent because of the success of support vector machines (SVM) in many classification tasks. We describe CRAIG, a new program for ab initio gene prediction based on a conditional random field model with semi-Markov structure that is trained with an online large-margin algorithm related to multiclass SVMs. Our experiments on benchmark vertebrate datasets and on regions from the ENCODE project show significant improvements in prediction accuracy over published gene predictors that use intrinsic features only, particularly at the gene level and on genes with long introns.     

Using multiple alignments to improve gene prediction.

The multiple species de novo gene prediction problem can be stated as follows: given an alignment of genomic sequences from two or more organisms, predict the location and structure of all protein-coding genes in one or more of the sequences. Here, we present a new system, N-SCAN (a.k.a. TWINSCAN 3.0), for addressing this problem. N-SCAN can model the phylogenetic relationships between the aligned genome sequences, context dependent substitution rates, and insertions and deletions. An implementation of N-SCAN was created and used to generate predictions for the entire human genome and the genome of the fruit fly Drosophila melanogaster. Analyses of the predictions reveal that N-SCAN's accuracy in both human and fly exceeds that of all previously published whole-genome de novo gene predictors.     

Hundreds of Out-of-Frame Remodeled Gene Families in the Escherichia coli Pangenome

All genomes include gene families with very limited taxonomic distributions that potentially represent new genes and innovations in protein-coding sequence, raising questions on the origins of such genes. Some of these genes are hypothesized to have formed de novo, from noncoding sequences, and recent work has begun to elucidate the processes by which de novo gene formation can occur. A special case of de novo gene formation, overprinting, describes the origin of new genes from noncoding alternative reading frames of existing open reading frames (ORFs). We argue that additionally, out-of-frame gene fission/fusion events of alternative reading frames of ORFs and out-of-frame lateral gene transfers could contribute to the origin of new gene families. To demonstrate this, we developed an original pattern-search in sequence similarity networks, enhancing the use of these graphs, commonly used to detect in-frame remodeled genes. We applied this approach to gene families in 524 complete genomes of Escherichia coli. We identified 767 gene families whose evolutionary history likely included at least one out-of-frame remodeling event. These genes with out-of-frame components represent ∼2.5% of all genes in the E. coli pangenome, suggesting that alternative reading frames of existing ORFs can contribute to a significant proportion of de novo genes in bacteria.     

SCGPred： A Score-based Method for Gene Structure Prediction by Combining Multiple Sources of Evidence

Predicting protein-coding genes still remains a significant challenge. Although a variety of computational programs that use commonly machine learning methods have emerged, the accuracy of predictions remains a low level when implementing in large genomic sequences. Moreover, computational gene finding in newly se- quenced genomes is especially a difficult task due to the absence of a training set of abundant validated genes. Here we present a new gene-finding program, SCGPred, to improve the accuracy of prediction by combining multiple sources of evidence. SCGPred can perform both supervised method in previously well-studied genomes and unsupervised one in novel genomes. By testing with datasets composed of large DNA sequences from human and a novel genome of Ustilago maydi, SCGPred gains a significant improvement in comparison to the popular ab initio gene predictors. We also demonstrate that SCGPred can significantly improve prediction in novel genomes by combining several foreign gene finders with similarity alignments, which is superior to other unsupervised methods. Therefore, SCGPred can serve as an alternative gene-finding tool for newly sequenced eukaryotic genomes. The program is freely available at http：//bio.scu.edu.cn/SCGPred/.     

Comparative genomics approach to detecting split-coding regions in a low-coverage genome: lessons from the chimaera Callorhinchus milii (Holocephali, Chondrichthyes)

Recent development of deep sequencing technologies has facilitated de novo genome sequencing projects, now conducted even by individual laboratories. However, this will yield more and more genome sequences that are not well assembled, and will hinder thorough annotation when no closely related reference genome is available. One of the challenging issues is the identification of protein-coding sequences split into multiple unassembled genomic segments, which can confound orthology assignment and various laboratory experiments requiring the identification of individual genes. In this study, using the genome of a cartilaginous fish, Callorhinchus milii, as test case, we performed gene prediction using a model specifically trained for this genome. We implemented an algorithm, designated ESPRIT, to identify possible linkages between multiple protein-coding portions derived from a single genomic locus split into multiple unassembled genomic segments. We developed a validation framework based on an artificially fragmented human genome, improvements between early and recent mouse genome assemblies, comparison with experimentally validated sequences from GenBank, and phylogenetic analyses. Our strategy provided insights into practical solutions for efficient annotation of only partially sequenced (low-coverage) genomes. To our knowledge, our study is the first formulation of a method to link unassembled genomic segments based on proteomes of relatively distantly related species as references.     

Strand-loop-strand motifs: prediction of hairpins and diverging turns in proteins

Beta-sheet proteins have been particularly challenging for de novo structure prediction methods, which tend to pair adjacent beta-strands into beta-hairpins and produce overly local topologies. To remedy this problem and facilitate de novo prediction of beta-sheet protein structures, we have developed a neural network that classifies strand-loop-strand motifs by local hairpins and nonlocal diverging turns by using the amino acid sequence as input. The neural network is trained with a representative subset of the Protein Data Bank and achieves a prediction accuracy of 75.9 +/- 4.4% compared to a baseline prediction rate of 59.1%. Hairpins are predicted with an accuracy of 77.3 +/- 6.1%, diverging turns with an accuracy of 73.9 +/- 6.0%. Incorporation of the beta-hairpin/diverging turn classification into the ROSETTA de novo structure prediction method led to higher contact order models and somewhat improved tertiary structure predictions for a test set of 11 all-beta-proteins and 3 alphabeta-proteins. The beta-hairpin/diverging turn classification from amino acid sequences is available online for academic use (Meiler and Kuhn, 2003; www.jens-meiler.de/turnpred.html).     

Gene prediction and gene classes in Arabidopsis thaliana

Gene prediction methods for eukaryotic genomes still are not fully satisfying. One way to improve gene prediction accuracy, proven to be relevant for prokaryotes, is to consider more than one model of genes. Thus, we used our classification of Arabidopsis thaliana genes in two classes (CU(1) and CU(2)), previously delineated according to statistical features, in the GeneMark gene identification program. For each gene class, as well as for the two classes combined, a Markov model was developed (respectively, GM-CU(1), GM-CU(2) and GM-all) and then used on a test set of 168 genes to compare their respective efficiency. We concluded from this analysis that GM-CU(1) is more sensitive than GM-CU(2) which seems to be more specific to a gene type. Besides, GM-all does not give better results than GM-CU(1) and combining results from GM-CU(1) and GM-CU(2) greatly improve prediction efficiency in comparison with predictions made with GM-all only. Thus, this work confirms the necessity to consider more than one gene model for gene prediction in eukaryotic genomes, and to look for gene classes in order to build these models.     

高通量测序技术在农作物全基因组序列测定中的应用概览

近几年飞速发展的高通量测序技术(next generation sequencing,NGS)在生命科学研究的各个领域充分展现了其低成本、高通量和应用面广等优势。在现代农业生物技术领域,利用高通量测序技术,科学家们不仅能更经济而高效对农作物、模式植物或不同栽培品种进行深入的全基因组测序、重测序,也可以对成百上千的栽培品种进行高效而准确的遗传差异分析、分子标记分析、连锁图谱分析、表观遗传学分析、转录组分析,进而改进农作物的育种技术,加快新品种的育种研究。其中,获得农作物的全基因组序列是其他研究和分析的基础。本文通过介绍近年来发表的一些利用高通量测序技术进行的农作物全基因组测定和组装的工作,展示高通量测序技术在现代农业生物技术领域的广泛前景以及其建立起来的研究基础。     

真菌基因组de novo测序组装的方法与实践

真菌基因组较其他真核生物基因组结构简单,长度短,易于测序、组装与注释,因此真菌基因组是研究真核生物基因组的模型。为研究真菌基因组组装策略,本研究基于Illumina HiSeq测序平台对烟曲霉菌株An16007基因组测序,分别使用5种de novo组装软件ABySS、SOAP-denovo、Velvet、MaSuRCA和IDBA-UD组装基因组,然后通过Augustus软件进行基因预测,BUSCO软件评估组装结果。研究发现,5种组装软件对基因组组装结果不同,ABySS组装的基因组较其他4种组装软件具有更高的完整性和准确性,且预测的基因数量较高,因此,ABySS更适合本研究基因组的组装。本研究提供了真菌de novo测序、组装及组装质量评估的技术流程,为基因组<100 Mb的真菌或其他生物基因组的研究提供参考。     

Protein structure prediction assisted with sparse NMR data in CASP13

CASP13 has investigated the impact of sparse NMR data on the accuracy of protein structure prediction. NOESY and ¹⁵N-¹H residual dipolar coupling data, typical of that obtained for ¹⁵N,¹³C-enriched, perdeuterated proteins up to about 40 kDa, were simulated for 11 CASP13 targets ranging in size from 80 to 326 residues. For several targets, two prediction groups generated models that are more accurate than those produced using baseline methods. Real NMR data collected for a de novo designed protein were also provided to predictors, including one data set in which only backbone resonance assignments were available. Some NMR-assisted prediction groups also did very well with these data. CASP13 also assessed whether incorporation of sparse NMR data improves the accuracy of protein structure prediction relative to nonassisted regular methods. In most cases, incorporation of sparse, noisy NMR data results in models with higher accuracy. The best NMR-assisted models were also compared with the best regular predictions of any CASP13 group for the same target. For six of 13 targets, the most accurate model provided by any NMR-assisted prediction group was more accurate than the most accurate model provided by any regular prediction group; however, for the remaining seven targets, one or more regular prediction method provided a more accurate model than even the best NMR-assisted model. These results suggest a novel approach for protein structure determination, in which advanced prediction methods are first used to generate structural models, and sparse NMR data is then used to validate and/or refine these models.     

FragGeneScan: predicting genes in short and error-prone reads

The advances of next-generation sequencing technology have facilitated metagenomics research that attempts to determine directly the whole collection of genetic material within an environmental sample (i.e. the metagenome). Identification of genes directly from short reads has become an important yet challenging problem in annotating metagenomes, since the assembly of metagenomes is often not available. Gene predictors developed for whole genomes (e.g. Glimmer) and recently developed for metagenomic sequences (e.g. MetaGene) show a significant decrease in performance as the sequencing error rates increase, or as reads get shorter. We have developed a novel gene prediction method FragGeneScan, which combines sequencing error models and codon usages in a hidden Markov model to improve the prediction of protein-coding region in short reads. The performance of FragGeneScan was comparable to Glimmer and MetaGene for complete genomes. But for short reads, FragGeneScan consistently outperformed MetaGene (accuracy improved ∼62% for reads of 400 bases with 1% sequencing errors, and ∼18% for short reads of 100 bases that are error free). When applied to metagenomes, FragGeneScan recovered substantially more genes than MetaGene predicted (>90% of the genes identified by homology search), and many novel genes with no homologs in current protein sequence database.     

From structure prediction to genomic screens for novel non-coding RNAs

Non-coding RNAs (ncRNAs) are receiving more and more attention not only as an abundant class of genes, but also as regulatory structural elements (some located in mRNAs). A key feature of RNA function is its structure. Computational methods were developed early for folding and prediction of RNA structure with the aim of assisting in functional analysis. With the discovery of more and more ncRNAs, it has become clear that a large fraction of these are highly structured. Interestingly, a large part of the structure is comprised of regular Watson-Crick and GU wobble base pairs. This and the increased amount of available genomes have made it possible to employ structure-based methods for genomic screens. The field has moved from folding prediction of single sequences to computational screens for ncRNAs in genomic sequence using the RNA structure as the main characteristic feature. Whereas early methods focused on energy-directed folding of single sequences, comparative analysis based on structure preserving changes of base pairs has been efficient in improving accuracy, and today this constitutes a key component in genomic screens. Here, we cover the basic principles of RNA folding and touch upon some of the concepts in current methods that have been applied in genomic screens for de novo RNA structures in searches for novel ncRNA genes and regulatory RNA structure on mRNAs. We discuss the strengths and weaknesses of the different strategies and how they can complement each other.     

Efficient decoding algorithms for generalized hidden Markov model gene finders

Background

The Generalized Hidden Markov Model (GHMM) has proven a useful framework for the task of computational gene prediction in eukaryotic genomes, due to its flexibility and probabilistic underpinnings. As the focus of the gene finding community shifts toward the use of homology information to improve prediction accuracy, extensions to the basic GHMM model are being explored as possible ways to integrate this homology information into the prediction process. Particularly prominent among these extensions are those techniques which call for the simultaneous prediction of genes in two or more genomes at once, thereby increasing significantly the computational cost of prediction and highlighting the importance of speed and memory efficiency in the implementation of the underlying GHMM algorithms. Unfortunately, the task of implementing an efficient GHMM-based gene finder is already a nontrivial one, and it can be expected that this task will only grow more onerous as our models increase in complexity.

Results

As a first step toward addressing the implementation challenges of these next-generation systems, we describe in detail two software architectures for GHMM-based gene finders, one comprising the common array-based approach, and the other a highly optimized algorithm which requires significantly less memory while achieving virtually identical speed. We then show how both of these architectures can be accelerated by a factor of two by optimizing their content sensors. We finish with a brief illustration of the impact these optimizations have had on the feasibility of our new homology-based gene finder, TWAIN.

Conclusions

In describing a number of optimizations for GHMM-based gene finders and making available two complete open-source software systems embodying these methods, it is our hope that others will be more enabled to explore promising extensions to the GHMM framework, thereby improving the state-of-the-art in gene prediction techniques.     

Progress and challenges in high-resolution refinement of protein structure models

Achieving atomic level accuracy in de novo structure prediction presents a formidable challenge even in the context of protein models with correct topologies. High-resolution refinement is a fundamental test of force field accuracy and sampling methodology, and its limited success in both comparative modeling and de novo prediction contexts highlights the limitations of current approaches. We constructed four tests to identify bottlenecks in our current approach and to guide progress in this challenging area. The first three tests showed that idealized native structures are stable under our refinement simulation conditions and that the refinement protocol can significantly decrease the root mean square deviation (RMSD) of perturbed native structures. In the fourth test we applied the refinement protocol to de novo models and showed that accurate models could be identified based on their energies, and in several cases many of the buried side chains adopted native-like conformations. We also showed that the differences in backbone and side-chain conformations between the refined de novo models and the native structures are largely localized to loop regions and regions where the native structure has unusual features such as rare rotamers or atypical hydrogen bonding between beta-strands. The refined de novo models typically have higher energies than refined idealized native structures, indicating that sampling of local backbone conformations and side-chain packing arrangements in a condensed state is a primary obstacle.