RNA-Seq, also called "Whole Transcriptome Shotgun Sequencing"  ("WTSS") and dubbed "a revolutionary tool for transcriptomics" , refers to the use of High-throughput sequencing technologies to sequence cDNA in order to get information about a sample's RNA content, a technique that is quickly becoming invaluable in the study of diseases like cancer .
Thanks to the deep coverage and base level resolution provided by next-generation sequencing instruments, RNA-Seq provides researchers with efficient ways to measure transcriptome data experimentally, allowing them to get information such as how different alleles of a gene are expressed, detect post-transcriptional mutations or identifying gene fusions .
- 1 Introduction
- 2 Methods
- 3 Analysis
- 4 References
The introduction of Next-generation sequencing or High-throughput sequencing technologies opened new doors into the field of DNA sequencing, however as understanding of these technologies becomes more widespread and new tools are developed, new innovative ways of applying these technologies are being created .
Given High-throughput sequencing technologies' low requirements of nucleotide sequence product, together with its deep coverage and base-scale resolution, its use has expanded to the field of transcriptomics . Transcriptomics is an area of research characterizing the RNA transcribed from a particular genome under investigation. Although transcriptomes are more dynamic relative to genomic DNA, these molecules provide direct access to gene regulation and protein information. Sequencing transcriptomes is not a new idea. Various methods have been developed previously to directly determine cDNA sequences based mostly around traditional (and more expensive) Sanger sequencing, while others include methodologies such as Serial analysis of gene expression (SAGE), cap analysis gene expression (CAGE) and massively parallel signature sequencing (MPSS).
Transcriptome Sequencing (RNA-seq) can be done with a variety of platforms to test a smorgasbord of ideas and hypotheses. For example, using the Illumina (company) Genome Analyzer platform, recent applications include sequencing mammalian transcriptomes , ABI Solid Sequencing to profile stem cell transcriptomes  or Life Science's 454 Sequencing to discover SNPs in maize . Even though each platform has its technical individualities, the information gathered from each is of the same nature.
RNA Poly(A) Library
Creation of a library can change from platform to platform in high throughput sequencing , where each has several kits designed to build different types of libraries and adapting the resulting sequences to the specific requirements of their instruments.
However, due to the nature of the template being analyzed, there are commonalities within each technology. Frequently, in mRNA analysis the 3' polyadenylated (poly(A)) tail is targeted in order to ensure that coding RNA is separated from noncoding RNA. This can be accomplished simply with poly (T) oligos covalently attached to a given substrate. Presently many studies utilize magnetic beads for this step (; ) (Invitrogen, MACS mRNA Isolation kit). The Protocol Online website  provides a list of several protocols relating to mRNA isolation.
Studies including portions of the transcriptome outside poly(A) RNAs have shown that when using poly(T) magnetic beads, the flow-through RNA (non-poly(A) RNA) can yield important noncoding RNA gene discovery which would have otherwise gone unnoticed 
Also, since ribosomal RNA represents over 90% of the RNA within a given cell, studies have shown that its removal via probe hybridization (Invitrogen, RiboMinus Human/Mouse Transcriptome Isolation kit) increases the capacity to retrieve data from the remaining portion of the transcriptome.
The next step is reverse transcription. Due to the 5' bias of randomly primed-reverse transcription as well as secondary structures influencing primer binding sites , hydrolysis of RNA into 200-300 nucleotides prior to reverse transcription reduces both problems simultaneously. However, there are trade-offs with this method where although the overall body of the transcripts are efficiently converted to DNA, the 5' and 3' ends are less so. Depending on the aim of the study, researchers may choose to apply or ignore this step.
Once the cDNA is synthesized it can be further fragmented to reach the desired fragment length as specified in table 1. The template is now ready to be prepared for the desired sequencing method.
High-throughput sequencing technologies generate millions of short reads from a library of nucleotide sequences, whether they come from DNA, RNA, or a mixture, the sequencing mechanism of each platform does not vary. The most used technologies and some of their characteristics are shown in the following table
|454 Sequencing ||Illumina ||SOLiD |
|Sequencing Chemistry||Pyrosequencing||Polymerase-based sequence-by-synthesis||Ligation-based sequencing|
|Amplification approach||Emulsion PCR||Bridge amplification||Emulsion PCR|
|Paired end separation||3 kb||200 bp||3 kb|
|Mb per run||100 Mb||1300 Mb||3000 Mb|
|Time per paired end run||7 hours||4 days||5 days|
|Read length (update)||250 bp (400 bp)||32-40 bp (35-100 bp)||35 bp (35-50 bp)|
|Cost per run||$ 8,438 USD||$ 8,950 USD||$ 17,447 USD|
|Cost per Mb||$ 84.39 USD||$ 5.97 USD||$ 5.81 USD|
Table 1. Comparing metrics and performance of next-generation DNA sequencers 
Due to the small size of the short reads (for Illumina Genome Analyzer this can be around 42 bases) de novo assembly may be difficult (though some software does exist: Velvet (algorithm)), as there cannot be large overlaps between each read needed to easily reconstruct the original sequences, and the deep coverage makes the computing power to track all the possible alignments prohibitive . This can be somewhat overcome by having larger sequences obtained from the same sample using other techniques as Sanger Sequencing, and using larger reads as a "skeleton" or a "template" to help assemble reads in difficult regions (e.g. regions with repetitive sequences).
The recommended approach is that of aligning the millions of reads to a "reference genome". There are many tools available for aligning genomic reads to a reference genome (sequence alignment tools), however, special attention is needed when alignment of a transcriptome to a genome, mainly when dealing with genes having intronic regions.
As discussed above, the sequence libraries are created extracting mRNA using its poly(A) tail, which is added to the mRNA molecule post-transcriptionally and thus splicing has taken place. Therefore, the created library and the short reads obtained cannot come from intronic sequences and thus, when trying to align these short reads to a reference genome, only short reads aligning entirely inside exonic regions will be matched while short reads from exon-exon junction regions will not.
A possible method to work around this is to try to align the unaligned short reads using a proxy genome generated with known exonic sequences. This need not cover whole exons, only enough so that the short reads can match on both sides of the exon-exon junction with minimum overlap. The use of paired-end sequencing has been mentioned as a good solution to alignment problems, as besides giving longer length reads, it allows obtaining information in respect to the strand .
Several software packages exist for short read alignment, and recently specialized algorithms for transcriptome alignment have been developed, e.g. TopHat  and Cufflinks  .
The characterization of gene expression in cells via measurement of mRNA levels has long been of interest to researchers. Even though it has been shown that due to other post transcriptional gene regulation events (such as RNA interference) there is not a strong correlation between the abundance of mRNA and the related proteins , measuring mRNA concentration levels is still a useful tool in determining how the transcriptional machinery of the cell is affected in the presence of external signals (e.g. drug treatment), or how cells differ between a healthy state and a diseased state.
Prior to RNA-Seq, DNA microarrays were unchallenged as the experiment of choice for transcriptome analysis. Although many experiments are still using microarrays to generate exciting results where the amount of time to retrieve results for a given sample is shorter, intrinsic experimental limitations of microarrays seem to make RNA-Seq the method of choice. One important limitation is a prerequisite for sequence information in order to detect and ultimately evaluate transcripts . As research in the field of RNA-Seq is growing steadily with promising and consistent results, one must now consider, "Is this the beginning of the end for microarrays?" . Still, for many applications, microarrays are the method of choice. Not only are they 10-100 times cheaper when compared at the same resolution of accuracy (less than $100 per array for high-throughput applications), but RNA-Seq protocols still suffer from unknown biases such as those implied by the required ligation steps, and known biases where high abundance transcripts (such as from housekeeping genes) make up the majority of sequencing data (e.g. in some tissues 5% of the genes represent up to 75% of the reads sequenced).
Coverage as measure of expression
Expression can be deduced via RNA-Seq to the extent at which a sequence is retrieved. Transcriptome studies in Yeast  show that in this experimental setting, a fourfold coverage is required for amplicons to be classified and characterized as an expressed gene. When the transcriptome is fragmented prior to cDNA synthesis, the number of reads corresponding to the particular exon normalized by its length in vivo yields gene expression levels which correlate with those obtained through qPCR.
Single nucleotide variation discovery
Transcriptome single nucleotide variation has been analyzed in maize on the Roche 454 sequencing platform . Directly from the transcriptome anaysis, around 7000 single nucleotide polymorphisms (SNPs) were recognized. Following Sanger sequence validation, the researchers were able to conservatively obtain almost 5000 valid SNPs covering more than 2400 maize genes. This impressive transcriptome analysis is currently being applied to cancer research and microbiology which could quite possibly lead to new forms of medicine.
Coverage/depth can affect the mutations seen and given that everything is expression-centric, an allele might not be detected either because it is not in the genome, or because it is not being expressed. At the same time, RNA-seq can yield additional information rather than just the existence of a heterozygous gene as it can also help in estimating the expression of each allele. In association studies, genotypes are associated to disease and expression levels can also be associated with disease. Using RNA-seq, we can measure the relationship between these two associated variables, that is, in what relation are each of the alleles being expressed.
The depth of sequencing required for specific applications can be extrapolated from a pilot experiment. 
Germline vs expressed alleles
The only way to be absolutely sure of the individual's mutations is to compare the transcriptome sequences to the germline DNA sequence. This enables the distinction of homozygous genes versus skewed expression of one of the alleles and it can also provide information about genes that were not expressed in the transcriptomic experiment.
Having the matching genomic and transcriptomic sequences of an individual can also help in detecting post-transcriptional edits , where, if the individual is homozygous for a gene, but the gene's transcript has a different allele, then a post-transcriptional modification event is determined.
mRNA centric single nucleotide variants (SNVs) are generally not considered as a representative source of functional variation in cells, mainly due to the fact that these mutations disappear with the mRNA molecule, however the fact that efficient DNA correction mechanisms do not apply to RNA molecules can cause them to appear more often. This has been proposed as the source of certain prion diseases , also known as TSE or transmissible spongiform encephalopathies.
Fusion gene detection
Caused by different structural modifications in the genome, fusion genes have gained attention because of their relationship with cancer . RNA-Seq's ability to analyze a sample's whole transcriptome in an unbiased fashion makes it an attractive tool to find these kinds of common events in cancer .
The idea follows from the process of aligning the short transcriptomic reads to a reference genome. Most of the short reads will fall within one complete exon, and a smaller but still large set would be expected to map to known exon-exon junctions. The remaining unmapped short reads would then be further analyzed to determine whether they match an exon-exon junction where the exons come from different genes. This would be evidence of a possible fusion event, however, because the length of the reads, this could prove to be very noisy. An alternative approach is using pair-end reads, when potentially a large number of paired reads would map each end to a different exon, giving better coverage of these events (see figure). Nonetheless, the end result consists of multiple and potentially novel combinations genes providing an ideal starting point for further validation.
The information gathered when sequencing a sample's transcriptome in this way has many of the same limitations as other RNA expression analysis pipelines. Mainly, the information gathered is:
a) Tissue specific: Gene expression is not uniform throughout an organism's cells, it is strongly dependent on the tissue type being measured;
b) Time dependent: During a cell's lifetime and context, its gene expression levels change.
Because of this, care must be taken when drawing conclusions from the sequencing experiment, as some of the information gathered might not be representative of the individual itself.
An example of this would be during SNV discovery as the mutations discovered are more precisely the mutations being expressed, this is: observing a homozygote location to a non-reference allele in an organism does not necessarily mean that this is the individual's genotype, it could just mean that the gene copy with the reference allele is not being expressed in that tissue and/or at the time snapshot the sample was acquired.
- ^ a b c Ryan D. Morin, Matthew Bainbridge, Anthony Fejes, Martin Hirst, Martin Krzywinski, Trevor J. Pugh, Helen McDonald, Richard Varhol, Steven J.M. Jones, and Marco A. Marra. (2008). "Profiling the HeLa S3 transcriptome using randomly primed cDNA and massively parallel short-read sequencing". BioTechniques 45 (1): 81–94. doi:10.2144/000112900. PMID 18611170. http://www.bcgsc.ca/about/pubann/biotechniques-publication-2008-44-8.
- ^ a b c d e Wang Z, Gerstein M, Snyder M. (January 2009). "RNA-Seq: a revolutionary tool for transcriptomics". Nature Reviews Genetics 10 (1): 57–63. doi:10.1038/nrg2484. PMID 19015660. http://www.nature.com/nrg/journal/v10/n1/abs/nrg2484.html.
- ^ a b c Maher CA, Kumar-Sinha C, Cao X, Kalyana-Sundaram S, Han B, Jing X, Sam L, Barrette T, Palanisamy N, Chinnaiyan AM (January 2009). "Transcriptome sequencing to detect gene fusions in cancer". Nature 458 (7234): 97–101. doi:10.1038/nature07638. PMID 19136943. PMC 2725402. http://www.nature.com/nature/journal/vaop/ncurrent/abs/nature07638.html.
- ^ Samuel Marguerat, Brian T. Wilhelm and Jürg Bähler (2008). "Next-generation sequencing: applications beyond genomes". Biochemical Society Transactions 36 (Pt 5): 1091–1096. doi:10.1042/BST0361091. PMID 18793195. PMC 2563889. http://www.biochemsoctrans.org/bst/036/1091/bst0361091.htm.
- ^ a b c d Mortazavi A, Williams BA, McCue K, Schaeffer L, Wold B. (2008). "Mapping and quantifying mammalian transcriptomes by RNA-Seq". Nature Methods 5 (7): 621–628. doi:10.1038/nmeth.1226. PMID 18516045. http://www.nature.com/nmeth/journal/v5/n7/abs/nmeth.1226.html.
- ^ Cloonan N, Forrest AR, Kolle G, Gardiner BB, Faulkner GJ, Brown MK, Taylor DF, Steptoe AL, Wani S, Bethel G, Robertson AJ, Perkins AC, Bruce SJ, Lee CC, Ranade SS, Peckham HE, Manning JM, McKernan KJ, Grimmond SM. (2008). "Stem cell transcriptome profiling via massive-scale mRNA sequencing". Nature Methods 5 (7): 613–619. doi:10.1038/nmeth.1223. PMID 18516046. http://www.nature.com/nmeth/journal/v5/n7/abs/nmeth.1223.html.
- ^ a b Barbazuk WB, Emrich SJ, Chen HD, Li L, Schnable PS (2007). "SNP discovery via 454 transcriptome sequencing". The Plant Journal 51 (5): 910–918. doi:10.1111/j.1365-313X.2007.03193.x. PMID 17662031. PMC 2169515. http://www3.interscience.wiley.com/journal/118488674/abstract.
- ^ http://www.protocol-online.org/prot/Molecular_Biology/RNA/RNA_Extraction/mRNA_Isolation/index.html
- ^ a b Mardis, ER (2008). "The impact of next-generation sequencing technology on genetics". Trends in Genetics 24 (3): 142–149. doi:10.1016/j.tig.2007.12.007. PMID 18262675.
- ^ http://www.454.com/applications/transcriptome-sequencing.asp
- ^ http://www.illumina.com/pages.ilmn?ID=204
- ^ http://solid.appliedbiosystems.com/
- ^ Zerbino DR, Birney E (2008). "Velvet: Algorithms for de novo short read assemblyusing de Bruijn graphs". Genome Research 18 (5): 821–829. doi:10.1101/gr.074492.107. PMID 18349386. PMC 2336801. http://genome.cshlp.org/content/18/5/821.full.
- ^ Cole Trapnell, Lior Pachter and Steven Sazlberg (2009). "TopHat: discovering splice junctions with RNA-Seq". Bioinformatics 25 (9): 1105–1111. doi:10.1093/bioinformatics/btp120. PMID 19289445. PMC 2672628. http://bioinformatics.oxfordjournals.org/cgi/content/abstract/25/9/1105?etoc.
- ^ Cole Trapnell, Brian A Williams, Geo Pertea, Ali Mortazavi, Gordon Kwan, Marijke J van Baren, Steven L Salzberg, Barbara J Wold and Lior Pachter (2010). "Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation". Nature Biotechnology 28 (5): 511–515. doi:10.1038/nbt.1621. PMID 20436464. http://www.nature.com/nbt/journal/v28/n5/abs/nbt.1621.html.
- ^ Greenbaum D, Colangelo C, Williams K, Gerstein M. (2003). "Comparing protein abundance and mRNA expression levels on a genomic scale". Genome Biology 4 (9): 117. doi:10.1186/gb-2003-4-9-117. PMID 12952525. PMC 193646. http://genomebiology.com/2003/4/9/117.
- ^ Marioni JC, Mason CE, Mane SM, Stephens M, Gilad Y. (2008). "RNA-seq: An assessment of technical reproducibility and comparison with gene expression arrays". Genome Research 18 (9): 1509–1517. doi:10.1101/gr.079558.108. PMID 18550803. PMC 2527709. http://genome.cshlp.org/content/18/9/1509.
- ^ Shendure, Jay (2008). "The beginning of the end for microarrays?". Nature Methods 5 (7): 585–587. doi:10.1038/nmeth0708-585. PMID 18587314. http://www.nature.com/nmeth/journal/v5/n7/abs/nmeth0708-585.html.
- ^ Nagalakshmi U, Wang Z, Waern K, Shou C, Raha D, Gerstein M, Snyder M (2008). "The transcriptional landscape of the yeast genome defined by RNA sequencing". Science 320 (5881): 1344–1349. doi:10.1126/science.1158441. PMID 18451266. http://www.sciencemag.org/cgi/content/abstract/320/5881/1344.
- ^ a b Li H, Lovci MT, Kwon YS, Rosenfeld MG, Fu XD, Yeo GW (2008). "Determination of tag density required for digital transcriptome analysis: application to an androgen-sensitive prostate cancer model.". Proc Natl Acad Sci USA 105 (51): 20179–84. doi:10.1073/pnas.0807121105. PMID 19088194. PMC 2603435. http://www.pnas.org/content/105/51/20179.long.
- ^ Garcion E, Wallace B, Pelletier L, Wion D. (2004). "RNA mutagenesis and sporadic prion diseases". Journal of Theoretical Biology 230 (2): 271–274. doi:10.1016/j.jtbi.2004.05.014. PMID 15302558.
- ^ Teixeira MR (2006). "Recurrent fusion oncogenes in carcinomas". Ciritical Reviews in Oncogenesis 12 (3-4): 257–271. PMID 17425505. http://www.begellhouse.com/journals/439f422d0783386a,1371844864dc630c,7499e7bf0e7ad511.html.