Indel detection from DNA and RNA sequencing data with transIndel

Relevance of mutation-derived neoantigens and non-classical antigens for anticancer therapies

Efficacy of cancer immunotherapies relies on correct recognition of tumor antigens by lymphocytes, eliciting thus functional responses capable of eliminating tumor cells. Therefore, important efforts have been carried out in antigen identification, with the aim of understanding mechanisms of response to immunotherapy and to design safer and more efficient strategies. In addition to classical tumor-associated antigens identified during the last decades, implementation of next-generation sequencing methodologies is enabling the identification of neoantigens (neoAgs) arising from mutations, leading to the development of new neoAg-directed therapies. Moreover, there are numerous non-classical tumor antigens originated from other sources and identified by new methodologies. Here, we review the relevance of neoAgs in different immunotherapies and the results obtained by applying neoAg-based strategies. In addition, the different types of non-classical tumor antigens and the best approaches for their identification are described. This will help to increase the spectrum of targetable molecules useful in cancer immunotherapies.

Development of Gene-Based InDel Markers on Putative Drought Stress-Responsive Genes and Genetic Diversity of Durian (Durio zibethinus)

Insertion-deletion (InDel) markers are co-dominant, relatively abundant and practical for agarose gel genotyping. InDel polymorphism usually affects gene functions. Nucleotide sequences of durian (Durio zibethinus) are available, but InDel makers have not been well established. This study aimed to develop drought-related gene-based InDel markers for durian through bioinformatic analysis of RNA-Seq datasets. The polymorphism of the markers was verified in 24 durian genotypes local to Thailand. Bioinformatic analysis indicated 496 InDel loci having lengths more than 9 bp. To evaluate these InDel markers, 15 InDel loci were selected. Nine markers were successfully amplified a clear polymorphic band pattern on 2% agarose gel. The polymorphic information content (PIC) of these nine markers ranged from 0.1103 to 0.5808. The genetic distance between the 24 genotypes ranged from 0.222 to 0.889. The phylogeny based on the nine InDel loci distinguished the 24 genotypes and divided samples into four groups. This set of gene-based InDel markers on putative drought-responsive genes will be useful for genetic studies.

Genetic basis of the historical iron-accumulating dgl and brz mutants in pea

The Pisum sativum (pea) mutants degenerate leaves (dgl) and bronze (brz) accumulate large amounts of iron in leaves. First described several decades ago, the two mutants have provided important insights into iron homeostasis in plants but the underlying mutations have remained unknown. Using exome sequencing we identified an in-frame deletion associated with dgl in a BRUTUS homolog. The deletion is absent from wild type and the original parent line. BRUTUS belongs to a small family of E3 ubiquitin ligases acting as negative regulators of iron uptake in plants. The brz mutation was previously mapped to chromosome 4, and superimposing this region to the pea genome sequence uncovered a mutation in OPT3, encoding an oligopeptide transporter with a plant-specific role in metal transport. The causal nature of the mutations was confirmed by additional genetic analyses. Identification of the mutated genes rationalizes many of the previously described phenotypes and provides new insights into shoot-to-root signaling of iron deficiency. Furthermore, the non-lethal mutations in these essential genes suggest new strategies for biofortification of crops with iron.

Mastering DNA chromatogram analysis in Sanger sequencing for reliable clinical analysis

BACKGROUND

Sanger dideoxy sequencing is vital in clinical analysis due to its accuracy, ability to analyze genetic markers like SNPs and STRs, capability to generate reliable DNA profiles, and its role in resolving complex clinical cases. The precision and robustness of Sanger sequencing contribute significantly to the scientific basis of clinical investigations. Though the reading of chromatograms seems to be a routine step, many errors conducted in PCR may lead to consequent limitations in the readings of AGCT peaks. These errors are possibly associated with improper DNA amplification and its subsequent interpretation of DNA sequencing files, such as noisy peaks, artifacts, and confusion between double-peak technical errors, heterozygosity, and double infection potentials. Thus, it is not feasible to read nucleic acid sequences without giving serious attention to these technical problems. To ensure the accuracy of DNA sequencing outcomes, it is also imperative to detect and rectify technical challenges that may lead to misinterpretation of the DNA sequence, resulting in errors and incongruities in subsequent analyses.

SHORT CONCLUSION

This overview sheds light on prominent technical concerns that can emerge prior to and during the interpretation of DNA chromatograms in Sanger sequencing, along with offering strategies to address them effectively. The significance of identifying and tackling these technical limitations during the chromatogram analysis is underscored in this review. Recognizing these concerns can aid in enhancing the quality of downstream analyses for Sanger sequencing results, which holds notable improvement in accuracy, reliability, and ability to provide crucial genetic information in clinical analysis.

ScanNeo2: a comprehensive workflow for neoantigen detection and immunogenicity prediction from diverse genomic and transcriptomic alterations

MOTIVATION

Neoantigens, tumor-specific protein fragments, are invaluable in cancer immunotherapy due to their ability to serve as targets for the immune system. Computational prediction of these neoantigens from sequencing data often requires multiple algorithms and sophisticated workflows, which are currently restricted to specific types of variants, such as single-nucleotide variants or insertions/deletions. Nevertheless, other sources of neoantigens are often overlooked.

RESULTS

We introduce ScanNeo2 an improved and fully automated bioinformatics pipeline designed for high-throughput neoantigen prediction from raw sequencing data. Unlike its predecessor, ScanNeo2 integrates multiple sources of somatic variants, including canonical- and exitron-splicing, gene fusion events, and various somatic variants. Our benchmark results demonstrate that ScanNeo2 accurately identifies neoantigens, providing a comprehensive and more efficient solution for neoantigen prediction.

AVAILABILITY AND IMPLEMENTATION

ScanNeo2 is freely available at https://github.com/ylab-hi/ScanNeo2/ and is accompanied by instruction and application data.

Multimodal CRISPR perturbations of GWAS loci associated with coronary artery disease in vascular endothelial cells

Genome-wide association studies have identified >250 genetic variants associated with coronary artery disease (CAD), but the causal variants, genes and molecular mechanisms remain unknown at most loci. We performed pooled CRISPR screens to test the impact of sequences at or near CAD-associated genetic variants on vascular endothelial cell functions. Using CRISPR knockout, inhibition and activation, we targeted 1998 variants at 83 CAD loci to assess their effect on three adhesion proteins (E-selectin, ICAM1, VCAM1) and three key endothelial functions (nitric oxide and reactive oxygen species production, calcium signalling). At a false discovery rate ≤10%, we identified significant CRISPR perturbations near 42 variants located within 26 CAD loci. We used base editing to validate a putative causal variant in the promoter of the FES gene. Although a few of the loci include genes previously characterized in endothelial cells (e.g. AIDA, ARHGEF26, ADAMTS7), most are implicated in endothelial dysfunction for the first time. Detailed characterization of one of these new loci implicated the RNA helicase DHX38 in vascular endothelial cell senescence. While promising, our results also highlighted several limitations in using CRISPR perturbations to functionally dissect GWAS loci, including an unknown false negative rate and potential off-target effects.

Identification of a large intra-exonic deletion in BRCA2 exon 18 in a pancreatic ductal adenocarcinoma

By 2030, pancreatic cancer will become the second leading cause of cancer-related deaths in the United States and in Europe. The management of patients with advanced pancreatic cancer relies on chemotherapy and poly (ADP-ribose) polymerase inhibitors for patients who carry BRCA1/2 inactivating alterations. Some variants, such as large insertion/deletions (Indels), inactivating BRCA1/2 and therefore of clinical relevance can be hard to detect by next-generation sequencing techniques. Here we report a 47-year-old patient presenting with pancreatic cancer whose tumour harbours a large somatic intra-exonic deletion of BRCA2 of 141 bp. This BRCA2 deletion, located in the C-terminal domain, can be considered as pathogenic and consequently affect tumorigenesis because it is involved in the interaction between the DSS1 protein and DNA. Thanks to the optimized bioinformatics algorithm, this intermediate size deletion in BRCA2 was identified, enabling personalized patient management via the inclusion of the patients in a clinical trial.

LPS-Induced Liver Injury of Magang Geese through Toll-like Receptor and MAPK Signaling Pathway

Lipopolysaccharide (LPS) is one of the main virulence factors of Gram-negative bacteria. In the process of waterfowl breeding, an inflammatory reaction due to LPS infection is easily produced, which leads to a decline in waterfowl performance. The liver plays a vital role in the immune response and the removal of toxic components. Therefore, it is necessary to study the mechanism of liver injury induced by LPS in goose. In this study, a total of 100 1-day-old goslings were randomly divided into a control group and LPS group after 3 days of pre-feeding. On days 21, 23, and 25 of the formal experiment, the control group was intraperitoneally injected with 0.5 mL normal saline, and the LPS group was intraperitoneally injected with LPS 2 mg/(kg body weight) once a day. On day 25 of the experiment, liver samples were collected 3 h after the injection of saline and LPS. The results of histopathology and biochemical indexes showed that the livers of the LPS group had liver morphological structure destruction and inflammatory cell infiltration, and the levels of ALT and AST were increased. Next, RNA sequencing analysis was used to determine the abundances and characteristics of the transcripts, as well as the associated somatic mutations and alternative splicing. We screened 727 differentially expressed genes (DEGs) with p < 0.05 and |log2(Fold Change)| ≥ 1, as the thresholds; GO and KEGG enrichment analysis showed that LPS-induced liver injury may be involved in the Toll-like receptor signaling pathway, MAPK signaling pathway, NOD-like receptor signaling pathway, FoxO, and PPAR signaling pathway. Finally, we intersected the genes enriched in the key pathway of LPS-induced liver injury with the top 50 key genes in protein−protein interaction networks to obtain 28 more critical genes. Among them, 17 genes were enriched in Toll-like signaling pathway and MAPK signaling pathway. Therefore, these results suggest that LPS-induced liver injury in geese may be the result of the joint action of Toll-like receptor, MAPK, NOD-like receptor, FoxO, and PPAR signaling pathway. Among them, the TLR7-mediated MAPK signaling pathway plays a major role.

Landscape of exitrons in gastric cancer

Background

Exitron is a new type of non-canonical alternative splicing. Accumulating evidence implies exitron may have pathological function and contribute to another source of anti-tumor immunogenicity in various cancers. Its role in gastric cancer remains poorly understood. Large-scale, multi-omics analysis could comprehensively characterize the landscape of exitrons in gastric cancer, reveal undiscovered mechanism and hopefully identify molecular biomarkers for predicting immunotherapy response.

Methods

We collected datasets from five studies for analysis. RNA sequencing was used for exitron identification. Somatic mutations were detected by whole exome sequencing. Neopeptides were confirmed by proteome mass spectrometry.

Findings

42174 gastric cancer-specific exitrons (GCSEs) were identified in 632 patients. GCSEs were clinically relevant to gender, age, Lauren type, tumor stage and prognosis. Tissue specificity test and pathogenic exitron prediction revealed their unique functional impact. GCSEs were mutually exclusive with mutations and demonstrated both unique and complementary function against TP53 mutation in gastric cancer. We further established splicing regulatory network to reveal upstream regulation of exitron splicing. We also evaluated the immunogenicity and diagnostic potential of GCSEs. Evidence of GCSEs-derived neopeptide expression was validated by whole proteome mass spectrometry. PD-1 and Siglecs were significantly increased in high neoantigen load patients. But exitron-related biomarkers failed to predict immunotherapy response, possibly due to small sample size and insufficient sequencing depth.

Interpretation

The present study provided a comprehensive multidimensional landscape of gastric cancer exitrons and underscores insights into underexplored mechanism in gastric cancer pathology.

Funding

The Guangdong Provincial Key Laboratory of Precision Medicine for Gastroinstestinal Cancer (2020B121201004), the Guangdong Provincial Major Talents Project (No. 2019JC05Y361) and National Natural Science Foundation of China (grant number:82172960 and 81872013).

FOXA1 in prostate cancer

Most prostate cancers initially respond to androgen deprivation therapy (ADT). With the long-term application of ADT, localized prostate cancer will progress to castration-resistant prostate cancer (CRPC), metastatic CRPC (mCRPC), and neuroendocrine prostate cancer (NEPC), and the transcriptional network shifted. Forkhead box protein A1 (FOXA1) may play a key role in this process through multiple mechanisms. To better understand the role of FOXA1 in prostate cancer, we review the interplay among FOXA1-targeted genes, modulators of FOXA1, and FOXA1 with a particular emphasis on androgen receptor (AR) function. Furthermore, we discuss the distinct role of FOXA1 mutations in prostate cancer and clinical significance of FOXA1. We summarize possible regulation pathways of FOXA1 in different stages of prostate cancer. We focus on links between FOXA1 and AR, which may play different roles in various types of prostate cancer. Finally, we discuss FOXA1 mutation and its clinical significance in prostate cancer. FOXA1 regulates the development of prostate cancer through various pathways, and it could be a biomarker for mCRPC and NEPC. Future efforts need to focus on mechanisms underlying mutation of FOXA1 in advanced prostate cancer. We believe that FOXA1 would be a prognostic marker and therapeutic target in prostate cancer.

Detecting Medium and Large Insertions and Deletions with transIndel

Insertions and deletions (indels) are primarily detected from DNA sequencing (DNA-seq) data, but their transcriptional consequences remain unexplored due to challenges in distinguishing medium- and large-sized indels from RNA splicing events in RNA-seq data. We introduce transIndel, a splice-aware algorithm that parses the chimeric alignments predicted by a short read aligner and reconstructs the mid-sized insertions and large deletions based on the linear alignments of split reads from DNA-seq or RNA-seq data. Here, we describe the method and provide a tutorial on the installation and application of transIndel.

Single-cell transcriptomic characterization of a gastrulating human embryo

Gastrulation is the fundamental process in all multicellular animals through which the basic body plan is first laid down1-4. It is pivotal in generating cellular diversity coordinated with spatial patterning. In humans, gastrulation occurs in the third week after fertilization. Our understanding of this process in humans is relatively limited and based primarily on historical specimens5-8, experimental models9-12 or, more recently, in vitro cultured samples13-16. Here we characterize in a spatially resolved manner the single-cell transcriptional profile of an entire gastrulating human embryo, staged to be between 16 and 19 days after fertilization. We use these data to analyse the cell types present and to make comparisons with other model systems. In addition to pluripotent epiblast, we identified primordial germ cells, red blood cells and various mesodermal and endodermal cell types. This dataset offers a unique glimpse into a central but inaccessible stage of our development. This characterization provides new context for interpreting experiments in other model systems and represents a valuable resource for guiding directed differentiation of human cells in vitro.

Vulcan: Improved long-read mapping and structural variant calling via dual-mode alignment

BACKGROUND

Long-read sequencing has enabled unprecedented surveys of structural variation across the entire human genome. To maximize the potential of long-read sequencing in this context, novel mapping methods have emerged that have primarily focused on either speed or accuracy. Various heuristics and scoring schemas have been implemented in widely used read mappers (minimap2 and NGMLR) to optimize for speed or accuracy, which have variable performance across different genomic regions and for specific structural variants. Our hypothesis is that constraining read mapping to the use of a single gap penalty across distinct mutational hot spots reduces read alignment accuracy and impedes structural variant detection.

FINDINGS

We tested our hypothesis by implementing a read-mapping pipeline called Vulcan that uses two distinct gap penalty modes, which we refer to as dual-mode alignment. The high-level idea is that Vulcan leverages the computed normalized edit distance of the mapped reads via minimap2 to identify poorly aligned reads and realigns them using the more accurate yet computationally more expensive long-read mapper (NGMLR). In support of our hypothesis, we show that Vulcan improves the alignments for Oxford Nanopore Technology long reads for both simulated and real datasets. These improvements, in turn, lead to improved accuracy for structural variant calling performance on human genome datasets compared to either of the read-mapping methods alone.

CONCLUSIONS

Vulcan is the first long-read mapping framework that combines two distinct gap penalty modes for improved structural variant recall and precision. Vulcan is open-source and available under the MIT License at https://gitlab.com/treangenlab/vulcan.

Integrated protocol for exitron and exitron-derived neoantigen identification using human RNA-seq data with ScanExitron and ScanNeo

Exitron splicing (EIS) events in cancers can disrupt functional protein domains to cause cancer driver effects. EIS has been recognized as a new source of tumor neoantigens. Here, we describe an integrated protocol for EIS and EIS-derived neoantigen identification using RNA-seq data. The protocol constitutes a step-by-step guide from data collection to neoantigen prediction.

For complete details on the use and execution of this protocol, please refer to Wang et al. (2021).

•

A protocol for identifying exitron and exitron-derived neoantigens

•

Special focus on data preparation, and troubleshooting

•

Optional steps for applying this protocol to analyze TCGA PRAD cancer cohort

Demystifying emerging bulk RNA-Seq applications: the application and utility of bioinformatic methodology

Significant innovations in next-generation sequencing techniques and bioinformatics tools have impacted our appreciation and understanding of RNA. Practical RNA sequencing (RNA-Seq) applications have evolved in conjunction with sequence technology and bioinformatic tools advances. In most projects, bulk RNA-Seq data is used to measure gene expression patterns, isoform expression, alternative splicing and single-nucleotide polymorphisms. However, RNA-Seq holds far more hidden biological information including details of copy number alteration, microbial contamination, transposable elements, cell type (deconvolution) and the presence of neoantigens. Recent novel and advanced bioinformatic algorithms developed the capacity to retrieve this information from bulk RNA-Seq data, thus broadening its scope. The focus of this review is to comprehend the emerging bulk RNA-Seq-based analyses, emphasizing less familiar and underused applications. In doing so, we highlight the power of bulk RNA-Seq in providing biological insights.

A pan-cancer transcriptome analysis of exitron splicing identifies novel cancer driver genes and neoepitopes

Exitron splicing (EIS) creates a cryptic intron (called an exitron) within a protein-coding exon to increase proteome diversity. EIS is poorly characterized, but emerging evidence suggests a role for EIS in cancer. Through a systematic investigation of EIS across 33 cancers from 9,599 tumor transcriptomes, we discovered that EIS affected 63% of human coding genes and that 95% of those events were tumor specific. Notably, we observed a mutually exclusive pattern between EIS and somatic mutations in their affected genes. Functionally, we discovered that EIS altered known and novel cancer driver genes for causing gain- or loss-of-function, which promotes tumor progression. Importantly, we identified EIS-derived neoepitopes that bind to major histocompatibility complex (MHC) class I or II. Analysis of clinical data from a clear cell renal cell carcinoma cohort revealed an association between EIS-derived neoantigen load and checkpoint inhibitor response. Our findings establish the importance of considering EIS alterations when nominating cancer driver events and neoantigens.

Neoantigen prediction in human breast cancer using RNA sequencing data

Neoantigens have attracted attention as biomarkers or therapeutic targets. However, accurate prediction of neoantigens is still challenging, especially in terms of its accuracy and cost. Variant detection using RNA sequencing (RNA‐seq) data has been reported to be a low‐accuracy but cost‐effective tool, but the feasibility of RNA‐seq data for neoantigen prediction has not been fully examined. In the present study, we used whole‐exome sequencing (WES) and RNA‐seq data of tumor and matched normal samples from six breast cancer patients to evaluate the utility of RNA‐seq data instead of WES data in variant calling to detect neoantigen candidates. Somatic variants were called in three protocols using: (i) tumor and normal WES data (DNA method, Dm); (ii) tumor and normal RNA‐seq data (RNA method, Rm); and (iii) combination of tumor RNA‐seq and normal WES data (Combination method, Cm). We found that the Rm had both high false‐positive and high false‐negative rates because this method depended greatly on the expression status of normal transcripts. When we compared the results of Dm with those of Cm, only 14% of the neoantigen candidates detected in Dm were identified in Cm, but the majority of the missed candidates lacked coverage or variant allele reads in the tumor RNA. In contrast, about 70% of the neoepitope candidates with higher expression and rich mutant transcripts could be detected in Cm. Our results showed that Cm could be an efficient and a cost‐effective approach to predict highly expressed neoantigens in tumor samples.

Comparative Transcriptome Profiling of Skeletal Muscle from Black Muscovy Duck at Different Growth Stages Using RNA-seq

In China, the production for duck meat is second only to that of chicken, and the demand for duck meat is also increasing. However, there is still unclear on the internal mechanism of regulating skeletal muscle growth and development in duck. This study aimed to identity candidate genes related to growth of duck skeletal muscle and explore the potential regulatory mechanism. RNA-seq technology was used to compare the transcriptome of skeletal muscles in black Muscovy ducks at different developmental stages (day 17, 21, 27, 31, and 34 of embryos and postnatal 6-month-olds). The SNPs and InDels of black Muscovy ducks at different growth stages were mainly in “INTRON”, “SYNONYMOUS_CODING”, “UTR_3_PRIME”, and “DOWNSTREAM”. The average number of AS in each sample was 37,267, mainly concentrated in TSS and TTS. Besides, a total of 19 to 5377 DEGs were detected in each pairwise comparison. Functional analysis showed that the DEGs were mainly involved in the processes of cell growth, muscle development, and cellular activities (junction, migration, assembly, differentiation, and proliferation). Many of DEGs were well known to be related to growth of skeletal muscle in black Muscovy duck, such as MyoG, FBXO1, MEF2A, and FoxN2. KEGG pathway analysis identified that the DEGs were significantly enriched in the pathways related to the focal adhesion, MAPK signaling pathway and regulation of the actin cytoskeleton. Some DEGs assigned to these pathways were potential candidate genes inducing the difference in muscle growth among the developmental stages, such as FAF1, RGS8, GRB10, SMYD3, and TNNI2. Our study identified several genes and pathways that may participate in the regulation of skeletal muscle growth in black Muscovy duck. These results should serve as an important resource revealing the molecular basis of muscle growth and development in duck.

ScanITD: Detecting internal tandem duplication with robust variant allele frequency estimation

BACKGROUND

Internal tandem duplications (ITDs) are tandem duplications within coding exons and are important prognostic markers and drug targets for acute myeloid leukemia (AML). Next-generation sequencing has enabled the discovery of ITD at single-nucleotide resolution. ITD allele frequency is used in the risk stratification of patients with AML; higher ITD allele frequency is associated with poorer clinical outcomes. However, the ITD allele frequency data are often unavailable to treating physicians and the detection of ITDs with accurate variant allele frequency (VAF) estimation remains challenging for short-read sequencing.

RESULTS

Here we present the ScanITD approach, which performs a stepwise seed-and-realignment procedure for ITD detection with accurate VAF prediction. The evaluations on simulated and real data demonstrate that ScanITD outperforms 3 state-of-the-art ITD detectors, especially for VAF estimation. Importantly, ScanITD yields better accuracy than general-purpose structural variation callers for predicting ITD size range duplications.

CONCLUSIONS

ScanITD enables the accurate identification of ITDs with robust VAF estimation. ScanITD is written in Python and is open-source software that is freely accessible at https://github.com/ylab-hi/ScanITD.

Improved indel detection in DNA and RNA via realignment with ABRA2

MOTIVATION

Genomic variant detection from next-generation sequencing has become established as an extremely important component of research and clinical diagnoses in both cancer and Mendelian disorders. Insertions and deletions (indels) are a common source of variation and can frequently impact functionality, thus making their detection vitally important. While substantial effort has gone into detecting indels from DNA, there is still opportunity for improvement. Further, detection of indels from RNA-Seq data has largely been an afterthought and offers another critical area for variant detection.

RESULTS

We present here ABRA2, a redesign of the original ABRA implementation that offers support for realignment of both RNA and DNA short reads. The process results in improved accuracy and scalability including support for human whole genomes. Results demonstrate substantial improvement in indel detection for a variety of data types, including those that were not previously supported by ABRA. Further, ABRA2 results in broad improvements to variant calling accuracy across a wide range of post-processing workflows including whole genomes, targeted exomes and transcriptome sequencing.

AVAILABILITY AND IMPLEMENTATION

ABRA2 is implemented in a combination of Java and C/C++ and is freely available to all from: https://github.com/mozack/abra2.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

ScanNeo: identifying indel-derived neoantigens using RNA-Seq data

Summary

Insertion and deletion (indels) have been recognized as an important source generating tumor-specific mutant peptides (neoantigens). The focus of indel-derived neoantigen identification has been on leveraging DNA sequencing such as whole exome sequencing, with the effort of using RNA-seq less well explored. Here we present ScanNeo, a fast-streamlined computational pipeline for analyzing RNA-seq to predict neoepitopes derived from small to large-sized indels. We applied ScanNeo in a prostate cancer cell line and validated our predictions with matched mass spectrometry data. Finally, we demonstrated that indel neoantigens predicted from RNA-seq were associated with checkpoint inhibitor response in a cohort of melanoma patients.

Availability and implementation

ScanNeo is implemented in Python. It is freely accessible at the GitHub repository (https://github.com/ylab-hi/ScanNeo).

Supplementary information

Supplementary data are available at Bioinformatics online.