MAP-RSeq: Mayo Analysis Pipeline for RNA sequencing

Krishna R Kalari 0 Asha A Nair 0 Jaysheel D Bhavsar 0 Daniel R O'Brien 0 Jaime I Davila 0 Matthew A Bockol 0 Jinfu Nie 0 Xiaojia Tang 0 Saurabh Baheti 0 Jay B Doughty 0 Sumit Middha 0 Hugues Sicotte 0 Aubrey E Thompson 3 Yan W Asmann 2 Jean-Pierre A Kocher 0 1 0 Department of Health Sciences Research, Mayo Clinic , 200 First Street SW, Rochester, MN 55905 , USA 1 Present Address: Department of Health Sciences Research, Mayo Clinic , 200 First Street SW, Rochester, MN 55905 , USA 2 Department of Health Sciences Research, Mayo Clinic , 4500 San Pablo Road, Jacksonville, FL 32224 , USA 3 Department of Cancer Biology, Mayo Clinic , 4500 San Pablo Road, Jacksonville, FL 32224 , USA Background: Although the costs of next generation sequencing technology have decreased over the past years, there is still a lack of simple-to-use applications, for a comprehensive analysis of RNA sequencing data. There is no one-stop shop for transcriptomic genomics. We have developed MAP-RSeq, a comprehensive computational workflow that can be used for obtaining genomic features from transcriptomic sequencing data, for any genome. Results: For optimization of tools and parameters, MAP-RSeq was validated using both simulated and real datasets. MAP-RSeq workflow consists of six major modules such as alignment of reads, quality assessment of reads, gene expression assessment and exon read counting, identification of expressed single nucleotide variants (SNVs), detection of fusion transcripts, summarization of transcriptomics data and final report. This workflow is available for Human transcriptome analysis and can be easily adapted and used for other genomes. Several clinical and research projects at the Mayo Clinic have applied the MAP-RSeq workflow for RNA-Seq studies. The results from MAP-RSeq have thus far enabled clinicians and researchers to understand the transcriptomic landscape of diseases for better diagnosis and treatment of patients. Conclusions: Our software provides gene counts, exon counts, fusion candidates, expressed single nucleotide variants, mapping statistics, visualizations, and a detailed research data report for RNA-Seq. The workflow can be executed on a standalone virtual machine or on a parallel Sun Grid Engine cluster. The software can be downloaded from http://bioinformaticstools.mayo.edu/research/maprseq/. - Background Next generation sequencing (NGS) technology breakthroughs have allowed us to define the transcriptomic landscape for cancers and other diseases [1]. RNASequencing (RNA-Seq) is information-rich; it enables researchers to investigate a variety of genomic features, such as gene expression, characterization of novel transcripts, alternative splice sites, single nucleotide variants (SNVs), fusion transcripts, long non-coding RNAs, small insertions, and small deletions. Multiple alignment software packages are available for read alignment, quality control methods, gene expression and transcript quantification methods for RNA-Seq [2-5]. However, the majority of the RNA-Seq bioinformatics methods are focused only on the analysis of a few genomic features for downstream analysis [6-9]. At present there is no comprehensive RNA-Seq workflow that can simply be installed and used for multiple genomic feature analysis. At the Mayo Clinic, we have developed MAP-RSeq - a comprehensive computational workflow, to align, assess and report multiple genomic features from paired-end RNA-Seq data efficiently with a quick turnaround time. We have tested a variety of tools and methods to accurately estimate genomic features from RNA-Seq data. Best performing publically available bioinformatics tools along with parameter optimization were included in our workflow. As needed we have integrated in-house methods or tools to fill in the gaps. We have thoroughly investigated and compared the available tools and have optimized parameters to make the workflow run seamlessly for both virtual machine and cluster environments. Our software has been tested with paired-end sequencing reads from all Illumina platforms. Thus far, we have processed 1,535 Mayo Clinic samples using the MAP-RSeq workflow. The MAP-RSeq research reports for RNA-Seq data have enabled Mayo Clinic researchers and clinicians to exchange datasets and findings. Standardizing the workflow has allowed us to build a system that enables us to investigate across multiple studies within the Mayo Clinic. MAP-RSeq is a production application that allows researchers with minimal expertise in LINUX or Windows to install, analyze and interpret RNA-Seq data. Implementation MAP-RSeq uses a variety of freely available bioinformatics tools along with in-house developed methods using Perl, Python, R, and Java. MAP-RSeq is available in two versions. The first version is single threaded and runs on a virtual machine (VM). The VM version is straightforward to install. The second version is multi-threaded and is designed to run on a cluster environment. Virtual machine Virtual machine version of MAP-RSeq is available for download at the following URL [10]. This includes a sample dataset, references (limited to chromosome 22), and the complete MAP-RSeq workflow pre-installed. Virtual Box software (free for Windows, Mac, and Linux at [11]) needs to be installed in the host system. The system also needs to meet the following requirements: at least 4GB of physical memory, and at least 10GB of available disk. Although our sample data is only from Human Chromosome 22, this virtual machine can be extended to the entire human reference genome or to Table 1 MAP-RSeq installation and run time for QuickStart virtual machine Time to import into VM Run time with sample data (chr22 only) ~ 20 minutes to download on consumer grade internet ~10 minutes to download on consumer grade internet ~6 hours (mostly downloading and indexing references) Depends on the sample data used other species. However this requires allocating more memory (~16GB) than may be available on a typical desktop system and building the index references files for the species of interest. Tables 1 and 2 shows the install and run time metrics of MAP-RSeq in virtual machine and Linux environments respectively. For Table 2, we downloaded the breast cancer cell line data from CGHub [12] and randomly chose 4 million reads to run through the QuickStart VM. It took 6 hours for the MAP-RSeq workflow to complete. It did not exceed the 4GB memory limit, but did rely heavily on the swap space provided; making it run slower than if it would have had more physical memory available. Job profiling indicates that the system could have used 11GB of memory for such a sample. Sun grid engine MAP-RSeq requires four processing cores with a total of 16GB RAM to get optimal performance. It also requires 8GB of storage space for tools and reference file installation. For MAP-RSeq execution the following packages such as JAVA version 1.6.0_17 or higher, Perl version 5.10.0 or higher, Python version (...truncated)