TY - JOUR
T1 - Evaluating Methods for Isolating Total RNA and Predicting the Success of Sequencing Phylogenetically Diverse Plant Transcriptomes
AU - Johnson, Marc T.J.
AU - Carpenter, Eric J.
AU - Tian, Zhijian
AU - Bruskiewich, Richard
AU - Burris, Jason N.
AU - Carrigan, Charlotte T.
AU - Chase, Mark W.
AU - Clarke, Neil D.
AU - Covshoff, Sarah
AU - dePamphilis, Claude W.
AU - Edger, Patrick P.
AU - Goh, Falicia
AU - Graham, Sean
AU - Greiner, Stephan
AU - Hibberd, Julian M.
AU - Jordon-Thaden, Ingrid
AU - Kutchan, Toni M.
AU - Leebens-Mack, James
AU - Melkonian, Michael
AU - Miles, Nicholas
AU - Myburg, Henrietta
AU - Patterson, Jordan
AU - Pires, J. Chris
AU - Ralph, Paula
AU - Rolf, Megan
AU - Sage, Rowan F.
AU - Soltis, Douglas
AU - Soltis, Pamela
AU - Stevenson, Dennis
AU - Stewart, C. Neal
AU - Surek, Barbara
AU - Thomsen, Christina J.M.
AU - Villarreal, Juan Carlos
AU - Wu, Xiaolei
AU - Zhang, Yong
AU - Deyholos, Michael K.
AU - Wong, Gane Ka Shu
PY - 2012/11/21
Y1 - 2012/11/21
N2 - Next-generation sequencing plays a central role in the characterization and quantification of transcriptomes. Although numerous metrics are purported to quantify the quality of RNA, there have been no large-scale empirical evaluations of the major determinants of sequencing success. We used a combination of existing and newly developed methods to isolate total RNA from 1115 samples from 695 plant species in 324 families, which represents >900 million years of phylogenetic diversity from green algae through flowering plants, including many plants of economic importance. We then sequenced 629 of these samples on Illumina GAIIx and HiSeq platforms and performed a large comparative analysis to identify predictors of RNA quality and the diversity of putative genes (scaffolds) expressed within samples. Tissue types (e.g., leaf vs. flower) varied in RNA quality, sequencing depth and the number of scaffolds. Tissue age also influenced RNA quality but not the number of scaffolds ≥1000 bp. Overall, 36% of the variation in the number of scaffolds was explained by metrics of RNA integrity (RIN score), RNA purity (OD 260/230), sequencing platform (GAIIx vs HiSeq) and the amount of total RNA used for sequencing. However, our results show that the most commonly used measures of RNA quality (e.g., RIN) are weak predictors of the number of scaffolds because Illumina sequencing is robust to variation in RNA quality. These results provide novel insight into the methods that are most important in isolating high quality RNA for sequencing and assembling plant transcriptomes. The methods and recommendations provided here could increase the efficiency and decrease the cost of RNA sequencing for individual labs and genome centers.
AB - Next-generation sequencing plays a central role in the characterization and quantification of transcriptomes. Although numerous metrics are purported to quantify the quality of RNA, there have been no large-scale empirical evaluations of the major determinants of sequencing success. We used a combination of existing and newly developed methods to isolate total RNA from 1115 samples from 695 plant species in 324 families, which represents >900 million years of phylogenetic diversity from green algae through flowering plants, including many plants of economic importance. We then sequenced 629 of these samples on Illumina GAIIx and HiSeq platforms and performed a large comparative analysis to identify predictors of RNA quality and the diversity of putative genes (scaffolds) expressed within samples. Tissue types (e.g., leaf vs. flower) varied in RNA quality, sequencing depth and the number of scaffolds. Tissue age also influenced RNA quality but not the number of scaffolds ≥1000 bp. Overall, 36% of the variation in the number of scaffolds was explained by metrics of RNA integrity (RIN score), RNA purity (OD 260/230), sequencing platform (GAIIx vs HiSeq) and the amount of total RNA used for sequencing. However, our results show that the most commonly used measures of RNA quality (e.g., RIN) are weak predictors of the number of scaffolds because Illumina sequencing is robust to variation in RNA quality. These results provide novel insight into the methods that are most important in isolating high quality RNA for sequencing and assembling plant transcriptomes. The methods and recommendations provided here could increase the efficiency and decrease the cost of RNA sequencing for individual labs and genome centers.
UR - http://www.scopus.com/inward/record.url?scp=84869751747&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84869751747&partnerID=8YFLogxK
U2 - 10.1371/journal.pone.0050226
DO - 10.1371/journal.pone.0050226
M3 - Article
C2 - 23185583
AN - SCOPUS:84869751747
SN - 1932-6203
VL - 7
JO - PloS one
JF - PloS one
IS - 11
M1 - e50226
ER -