Transcript copy number estimation using a mouse whole-genome oligonucleotide microarray

Abstract

The ability to quantitatively measure the expression of all genes in a given tissue or cell with a single assay is an exciting promise of gene-expression profiling technology. An in situ-synthesized 60-mer oligonucleotide microarray designed to detect transcripts from all mouse genes was validated, as well as a set of exogenous RNA controls derived from the yeast genome (made freely available without restriction), which allow quantitative estimation of absolute endogenous transcript abundance.

Figure 1

60-mer oligonucleotide probe linearity testing. To test the performance of 21,044 60-mer oligonucleotide probes, E12.5 embryo RNA and placenta RNA were combined to form five pairs of duplicate samples containing from 0 to 100% placental RNA. Box-plot distribution data for each placental RNA input level is shown above, with median values labeled. The boxes show the 25-75 percentile range, with the mean and median indicated by the central straight line and diamond, respectively. Upper and lower bars show the 2.5 to 97.5 percentile range. Observed fraction medians are within 0.075 of input values, and 95% of values are within 0.405 of input values.

Figure 2

Relating yeast spike-in RNA control copy number to qPCR measurements and microarray signal intensity. (a) To verify abundances of yeast sequence RNA transcripts in a control mixture, cDNA was transcribed from the control mixture alone (open boxes), as well as E12.5 whole-mouse embryo total RNA (open diamonds) and Universal Mouse RNA (filled triangles) with added spike-in control mixture. The cDNA was used as template for real-time PCR quantitation of each yeast sequence RNA, using a separately prepared standard of cDNA transcribed from the yeast sequences. Expected and measured copy numbers are closely matched (r² ≥ 0.99), with maximum measured/observed ratios of 1.5, 1.5, and 2.6, respectively. (b) Expression profiles were generated for triplicate total RNA samples from E12.5 embryo (filled circles), E12.5 placenta (open circles), ES cells (filled boxes), and TS cells (open boxes) with yeast sequence control transcripts spiked-in prior to target labeling. For the seven control transcripts, mean log₁₀[intensity] is shown for each tissue type, as well as the mean across all samples (filled triangles), and these data were used to perform linear regression analysis and relate signal intensity to transcript copy number, allowing abundance estimation for endogenous transcripts. The regression line for the average of all tissues (dashed line) and its equation is shown. Intensity-copy number correlations for individual tissues were very strong, with r² values of 0.98 - 0.99.

Figure 3

Exogenous control and endogenous transcript amplification rates are closely matched over seven orders of magnitude. Transcript abundance of each spike-in control transcript was measured by qPCR before and after linear amplification labeling, and compared to amounts of the exogenous transcript Dnchc1. After amplification, individual ratios of each control transcript to the endogenous transcript were within 3.5-fold (average = 1.98-fold) of those prior to amplification. Blue diamonds = log₁₀[ratio mean control/Dnchc1 transcripts] of three E12.5 embryo and three E12.5 placenta samples before amplification. Red boxes, green triangles = log₁₀[ratio mean control/Dnchc1 transcripts] for the same samples after amplification, using yield versus input (red boxes) or the increase in Dnchc1 transcripts as measured by qPCR (green triangles) to calculate the fraction of the original sample represented by each qPCR well.

Figure 4

Validation of transcript abundance estimation for endogenous transcripts. qPCR primer sets were designed for selected genes so that amplicons were upstream of 60-mer oligonucleotide probes when possible, or less than 650 bp downstream, and copy number was estimated using serial dilutions of RNA, in vitro transcribed from mouse cDNAs, at known copy numbers as standards. Error bars represent one standard deviation across three replicate samples for each tissue. Dotted diagonal lines represent five- and tenfold differences between the two datasets. Each gene's official symbol, along with the unique identifier for the 60-mer oligonucleotide probe it was measured with, are listed in the key. Data was normalized to Gapd expression for both methods. EM = E12.5 embryo, PL = E12.5 placenta, ES = embryonic stem cells, TS = trophoblast stem cells.

Figure 5

Distribution of mouse transcript abundances in E12.5 embryo and placenta, and cultured ES and TS cells. Transcript abundances are expressed as log₁₀[copies per cell], varying over six orders of magnitude. The distributions are highly similar, despite the significant differences between the four tissues (for example, monolayer culture versus tissue, placenta versus embryo), suggesting that such distributions are not heavily skewed according to tissue structure or function. The percentage of transcripts present at less than one copy per cell ranged from 40.1 to 48.2% in the four tissues. Bins were centered on indicated values, and the dotted lines indicate values corresponding to mean upper and lower signal intensity reliability limits of one copy per 26 cells to 2,188 copies per cell. For definitions of tissue type see Figure 4 legend.