Support vector machines-based identification of alternative splicing in Arabidopsis thaliana from whole-genome tiling arrays

A Comprehensive Review of High Throughput Phenotyping and Machine Learning for Plant Stress Phenotyping

During the last decade, there has been rapid adoption of ground and aerial platforms with multiple sensors for phenotyping various biotic and abiotic stresses throughout the developmental stages of the crop plant. High throughput phenotyping (HTP) involves the application of these tools to phenotype the plants and can vary from ground-based imaging to aerial phenotyping to remote sensing. Adoption of these HTP tools has tried to reduce the phenotyping bottleneck in breeding programs and help to increase the pace of genetic gain. More specifically, several root phenotyping tools are discussed to study the plant's hidden half and an area long neglected. However, the use of these HTP technologies produces big data sets that impede the inference from those datasets. Machine learning and deep learning provide an alternative opportunity for the extraction of useful information for making conclusions. These are interdisciplinary approaches for data analysis using probability, statistics, classification, regression, decision theory, data visualization, and neural networks to relate information extracted with the phenotypes obtained. These techniques use feature extraction, identification, classification, and prediction criteria to identify pertinent data for use in plant breeding and pathology activities. This review focuses on the recent findings where machine learning and deep learning approaches have been used for plant stress phenotyping with data being collected using various HTP platforms. We have provided a comprehensive overview of different machine learning and deep learning tools available with their potential advantages and pitfalls. Overall, this review provides an avenue for studying various HTP platforms with particular emphasis on using the machine learning and deep learning tools for drawing legitimate conclusions. Finally, we propose the conceptual challenges being faced and provide insights on future perspectives for managing those issues.

Meta-analysis of drought-tolerant genotypes in Oryza sativa: A network-based approach

BACKGROUND

Drought is a severe environmental stress. It is estimated that about 50% of the world rice production is affected mainly by drought. Apart from conventional breeding strategies to develop drought-tolerant crops, innovative computational approaches may provide insights into the underlying molecular mechanisms of stress response and identify drought-responsive markers. Here we propose a network-based computational approach involving a meta-analytic study of seven drought-tolerant rice genotypes under drought stress.

RESULTS

Co-expression networks enable large-scale analysis of gene-pair associations and tightly coupled clusters that may represent coordinated biological processes. Considering differentially expressed genes in the co-expressed modules and supplementing external information such as resistance/tolerance QTLs, transcription factors, network-based topological measures, we identify and prioritize drought-adaptive co-expressed gene modules and potential candidate genes. Using the candidate genes that are well-represented across the datasets as 'seed' genes, two drought-specific protein-protein interaction networks (PPINs) are constructed with up- and down-regulated genes. Cluster analysis of the up-regulated PPIN revealed ABA signalling pathway as a central process in drought response with a probable crosstalk with energy metabolic processes. Tightly coupled gene clusters representing up-regulation of core cellular respiratory processes and enhanced degradation of branched chain amino acids and cell wall metabolism are identified. Cluster analysis of down-regulated PPIN provides a snapshot of major processes associated with photosynthesis, growth, development and protein synthesis, most of which are shut down during drought. Differential regulation of phytohormones, e.g., jasmonic acid, cell wall metabolism, signalling and posttranslational modifications associated with biotic stress are elucidated. Functional characterization of topologically important, drought-responsive uncharacterized genes that may play a role in important processes such as ABA signalling, calcium signalling, photosynthesis and cell wall metabolism is discussed. Further transgenic studies on these genes may help in elucidating their biological role under stress conditions.

CONCLUSION

Currently, a large number of resources for rice functional genomics exist which are mostly underutilized by the scientific community. In this study, a computational approach integrating information from various resources such as gene co-expression networks, protein-protein interactions and pathway-level information is proposed to provide a systems-level view of complex drought-responsive processes across the drought-tolerant genotypes.

50/50 Expressional Odds of Retention Signifies the Distinction between Retained Introns and Constitutively Spliced Introns in Arabidopsis thaliana

Intron retention, one of the most prevalent alternative splicing events in plants, can lead to introns retained in mature mRNAs. However, in comparison with constitutively spliced introns (CSIs), the relevantly distinguishable features for retained introns (RIs) are still poorly understood. This work proposes a computational pipeline to discover novel RIs from multiple next-generation RNA sequencing (RNA-Seq) datasets of Arabidopsis thaliana. Using this pipeline, we detected 3,472 novel RIs from 18 RNA-Seq datasets and re-confirmed 1,384 RIs which are currently annotated in the TAIR10 database. We also use the expression of intron-containing isoforms as a new feature in addition to the conventional features. Based on these features, RIs are highly distinguishable from CSIs by machine learning methods, especially when the expressional odds of retention (i.e., the expression ratio of the RI-containing isoforms relative to the isoforms without RIs for the same gene) reaches to or larger than 50/50. In this case, the RIs and CSIs can be clearly separated by the Random Forest with an outstanding performance of 0.95 on AUC (the area under a receiver operating characteristics curve). The closely related characteristics to the RIs include the low strength of splice sites, high similarity with the flanking exon sequences, low occurrence percentage of YTRAY near the acceptor site, existence of putative intronic splicing silencers (ISSs, i.e., AG/GA-rich motifs) and intronic splicing enhancers (ISEs, i.e., TTTT-containing motifs), and enrichment of Serine/Arginine-Rich (SR) proteins and heterogeneous nuclear ribonucleoparticle proteins (hnRNPs).

Computational Identification of Tissue-Specific Splicing Regulatory Elements in Human Genes from RNA-Seq Data

Alternative splicing is a vital process for regulating gene expression and promoting proteomic diversity. It plays a key role in tissue-specific expressed genes. This specificity is mainly regulated by splicing factors that bind to specific sequences called splicing regulatory elements (SREs). Here, we report a genome-wide analysis to study alternative splicing on multiple tissues, including brain, heart, liver, and muscle. We propose a pipeline to identify differential exons across tissues and hence tissue-specific SREs. In our pipeline, we utilize the DEXSeq package along with our previously reported algorithms. Utilizing the publicly available RNA-Seq data set from the Human BodyMap project, we identified 28,100 differentially used exons across the four tissues. We identified tissue-specific exonic splicing enhancers that overlap with various previously published experimental and computational databases. A complicated exonic enhancer regulatory network was revealed, where multiple exonic enhancers were found across multiple tissues while some were found only in specific tissues. Putative combinatorial exonic enhancers and silencers were discovered as well, which may be responsible for exon inclusion or exclusion across tissues. Some of the exonic enhancers are found to be co-occurring with multiple exonic silencers and vice versa, which demonstrates a complicated relationship between tissue-specific exonic enhancers and silencers.

SplAdder: identification, quantification and testing of alternative splicing events from RNA-Seq data

MOTIVATION

Understanding the occurrence and regulation of alternative splicing (AS) is a key task towards explaining the regulatory processes that shape the complex transcriptomes of higher eukaryotes. With the advent of high-throughput sequencing of RNA (RNA-Seq), the diversity of AS transcripts could be measured at an unprecedented depth. Although the catalog of known AS events has grown ever since, novel transcripts are commonly observed when working with less well annotated organisms, in the context of disease, or within large populations. Whereas an identification of complete transcripts is technically challenging and computationally expensive, focusing on single splicing events as a proxy for transcriptome characteristics is fruitful and sufficient for a wide range of analyses.

RESULTS

We present SplAdder, an alternative splicing toolbox, that takes RNA-Seq alignments and an annotation file as input to (i) augment the annotation based on RNA-Seq evidence, (ii) identify alternative splicing events present in the augmented annotation graph, (iii) quantify and confirm these events based on the RNA-Seq data and (iv) test for significant quantitative differences between samples. Thereby, our main focus lies on performance, accuracy and usability.

AVAILABILITY

Source code and documentation are available for download at http://github.com/ratschlab/spladder Example data, introductory information and a small tutorial are accessible via http://bioweb.me/spladder

CONTACTS

: andre.kahles@ratschlab.org or gunnar.ratsch@ratschlab.org

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Identifying splicing regulatory elements with de Bruijn graphs

Splicing regulatory elements (SREs) are short, degenerate sequences on pre-mRNA molecules that enhance or inhibit the splicing process via the binding of splicing factors, proteins that regulate the functioning of the spliceosome. Existing methods for identifying SREs in a genome are either experimental or computational. Here, we propose a formalism based on de Bruijn graphs that combines genomic structure, word count enrichment analysis, and experimental evidence to identify SREs found in exons. In our approach, SREs are not restricted to a fixed length (i.e., k-mers, for a fixed k). As a result, we identify 2001 putative exonic enhancers and 3080 putative exonic silencers for human genes, with lengths varying from 6 to 15 nucleotides. Many of the predicted SREs overlap with experimentally verified binding sites. Our model provides a novel method to predict variable length putative regulatory elements computationally for further experimental investigation.

Post-transcriptional and post-translational regulations of drought and heat response in plants: a spider's web of mechanisms

Drought and heat tolerance are complex quantitative traits. Moreover, the adaptive significance of some stress-related traits is more related to plant survival than to agronomic performance. A web of regulatory mechanisms fine-tunes the expression of stress-related traits and integrates both environmental and developmental signals. Both post-transcriptional and post-translational modifications contribute substantially to this network with a pivotal regulatory function of the transcriptional changes related to cellular and plant stress response. Alternative splicing and RNA-mediated silencing control the amount of specific transcripts, while ubiquitin and SUMO modify activity, sub-cellular localization and half-life of proteins. Interactions across these modification mechanisms ensure temporally and spatially appropriate patterns of downstream-gene expression. For key molecular components of these regulatory mechanisms, natural genetic diversity exists among genotypes with different behavior in terms of stress tolerance, with effects upon the expression of adaptive morphological and/or physiological target traits.

Comparative analyses between retained introns and constitutively spliced introns in Arabidopsis thaliana using random forest and support vector machine

One of the important modes of pre-mRNA post-transcriptional modification is alternative splicing. Alternative splicing allows creation of many distinct mature mRNA transcripts from a single gene by utilizing different splice sites. In plants like Arabidopsis thaliana, the most common type of alternative splicing is intron retention. Many studies in the past focus on positional distribution of retained introns (RIs) among different genic regions and their expression regulations, while little systematic classification of RIs from constitutively spliced introns (CSIs) has been conducted using machine learning approaches. We used random forest and support vector machine (SVM) with radial basis kernel function (RBF) to differentiate these two types of introns in Arabidopsis. By comparing coordinates of introns of all annotated mRNAs from TAIR10, we obtained our high-quality experimental data. To distinguish RIs from CSIs, We investigated the unique characteristics of RIs in comparison with CSIs and finally extracted 37 quantitative features: local and global nucleotide sequence features of introns, frequent motifs, the signal strength of splice sites, and the similarity between sequences of introns and their flanking regions. We demonstrated that our proposed feature extraction approach was more accurate in effectively classifying RIs from CSIs in comparison with other four approaches. The optimal penalty parameter C and the RBF kernel parameter in SVM were set based on particle swarm optimization algorithm (PSOSVM). Our classification performance showed F-Measure of 80.8% (random forest) and 77.4% (PSOSVM). Not only the basic sequence features and positional distribution characteristics of RIs were obtained, but also putative regulatory motifs in intron splicing were predicted based on our feature extraction approach. Clearly, our study will facilitate a better understanding of underlying mechanisms involved in intron retention.

Genome-wide analysis of alternative splicing of pre-mRNA under salt stress in Arabidopsis

Background

Alternative splicing (AS) of precursor mRNA (pre-mRNA) is an important gene regulation process that potentially regulates many physiological processes in plants, including the response to abiotic stresses such as salt stress.

Results

To analyze global changes in AS under salt stress, we obtained high-coverage (~200 times) RNA sequencing data from Arabidopsis thaliana seedlings that were treated with different concentrations of NaCl. We detected that ~49% of all intron-containing genes were alternatively spliced under salt stress, 10% of which experienced significant differential alternative splicing (DAS). Furthermore, AS increased significantly under salt stress compared with under unstressed conditions. We demonstrated that most DAS genes were not differentially regulated by salt stress, suggesting that AS may represent an independent layer of gene regulation in response to stress. Our analysis of functional categories suggested that DAS genes were associated with specific functional pathways, such as the pathways for the responses to stresses and RNA splicing. We revealed that serine/arginine-rich (SR) splicing factors were frequently and specifically regulated in AS under salt stresses, suggesting a complex loop in AS regulation for stress adaptation. We also showed that alternative splicing site selection (SS) occurred most frequently at 4 nucleotides upstream or downstream of the dominant sites and that exon skipping tended to link with alternative SS.

Conclusions

Our study provided a comprehensive view of AS under salt stress and revealed novel insights into the potential roles of AS in plant response to salt stress.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-15-431) contains supplementary material, which is available to authorized users.

A brief introduction to tiling microarrays: principles, concepts, and applications

Technological achievements have always contributed to the advancement of biomedical research. It has never been more so than in recent times, when the development and application of innovative cutting-edge technologies have transformed biology into a data-rich quantitative science. This stunning revolution in biology primarily ensued from the emergence of microarrays over two decades ago. The completion of whole-genome sequencing projects and the advance in microarray manufacturing technologies enabled the development of tiling microarrays, which gave unprecedented genomic coverage. Since their first description, several types of application of tiling arrays have emerged, each aiming to tackle a different biological problem. Although numerous algorithms have already been developed to analyze microarray data, new method development is still needed not only for better performance but also for integration of available microarray data sets, which without doubt constitute one of the largest collections of biological data ever generated. In this chapter we first introduce the principles behind the emergence and the development of tiling microarrays, and then discuss with some examples how they are used to investigate different biological problems.

Multiple insert size paired-end sequencing for deconvolution of complex transcriptomes

Deep sequencing of transcriptomes allows quantitative and qualitative analysis of many RNA species in a sample, with parallel comparison of expression levels, splicing variants, natural antisense transcripts, RNA editing and transcriptional start and stop sites the ideal goal. By computational modeling, we show how libraries of multiple insert sizes combined with strand-specific, paired-end (SS-PE) sequencing can increase the information gained on alternative splicing, especially in higher eukaryotes. Despite the benefits of gaining SS-PE data with paired ends of varying distance, the standard Illumina protocol allows only non-strand-specific, paired-end sequencing with a single insert size. Here, we modify the Illumina RNA ligation protocol to allow SS-PE sequencing by using a custom pre-adenylated 3' adaptor. We generate parallel libraries with differing insert sizes to aid deconvolution of alternative splicing events and to characterize the extent and distribution of natural antisense transcription in C. elegans. Despite stringent requirements for detection of alternative splicing, our data increases the number of intron retention and exon skipping events annotated in the Wormbase genome annotations by 127% and 121%, respectively. We show that parallel libraries with a range of insert sizes increase transcriptomic information gained by sequencing and that by current established benchmarks our protocol gives competitive results with respect to library quality.

Alternative splicing: enhancing ability to cope with stress via transcriptome plasticity

Alternative splicing is a mechanism for the regulation of gene expression that is widespread in higher eukaryotes. Genome-wide approaches, based on comparison of expressed and genomic sequences, on tiling arrays, and on next-generation sequencing, have provided growing knowledge of the extent, distribution and association of alternative splicing with stress-related genes in plants. The functional meaning of alternative splicing in response to stress has been defined for many genes, and in particular for those involved in the regulation of the stress responses, such as protein kinases, transcription factors, splicing regulators and pathogen-resistance genes. The production of proteins with diverse domain rearrangements from the same gene is the main alternative splicing mechanism for pathogen-resistance genes. The plant response to abiotic stress is also characterized by a second mechanism, which consists of the expression of alternative transcripts that are targeted to nonsense-mediated decay. These quantitatively regulate stress-related gene expression. Many alternative splicing events are well conserved among plant species, and also across kingdoms, especially those observed in response to stress, for genes encoding splicing regulators, and other classes of RNA-binding proteins. Nevertheless, non-conserved events indicate that alternative splicing represents an evolutionary strategy that rapidly increases genome plasticity and develops new gene functions, along with other mechanisms such as gene duplication. Finally, the study of the naturally occurring variability of alternative splicing and the identification of genomic regions involved in the regulation of alternative splicing in crops are proposed as strategies for selecting genotypes with superior performance under adverse environmental conditions.