A general framework for optimization of probes for gene expression microarray and its application to the fungus Podospora anserina

Abstract

Background: The development of new microarray technologies makes custom long oligonucleotide arrays affordable for many experimental applications, notably gene expression analyses. Reliable results depend on probe design quality and selection. Probe design strategy should cope with the limited accuracy of de novo gene prediction programs, and annotation up-dating. We present a novel in silico procedure which addresses these issues and includes experimental screening, as an empirical approach is the best strategy to identify optimal probes in the in silico outcome.

Findings: We used four criteria for in silico probe selection: cross-hybridization, hairpin stability, probe location relative to coding sequence end and intron position. This latter criterion is critical when exon-intron gene structure predictions for intron-rich genes are inaccurate. For each coding sequence (CDS), we selected a sub-set of four probes. These probes were included in a test microarray, which was used to evaluate the hybridization behavior of each probe. The best probe for each CDS was selected according to three experimental criteria: signal-to-noise ratio, signal reproducibility, and representative signal intensities. This procedure was applied for the development of a gene expression Agilent platform for the filamentous fungus Podospora anserina and the selection of a single 60-mer probe for each of the 10,556 P. anserina CDS.

Conclusions: A reliable gene expression microarray version based on the Agilent 44K platform was developed with four spot replicates of each probe to increase statistical significance of analysis.

Figure 1

Probe classes according to their position relative to an intron. The black arrow represents the coding strand of a gene. Probes are identical to the coding strand. Nucleotide numbering begins at the first nucleotide of the contig, on the coding strand of the gene of interest; x represents the numbering of the last nucleotide of the exon preceding the 5' end of the intron, and y represents the numbering of the first nucleotide following the 3' end. Probe classes are indicated by the colored boxes.

Figure 2

Flow diagram for probe selection.

Figure 3

Distribution of probe intensity CV in the five conditions used for the experimental validation of probes. The distributions of probe intensity CV are presented in a series of five boxes (interquartile range) and whiskers plots. Hybridizations were performed on microarray v.2 with the cRNAs prepared from the five conditions (M24h, M48h, M96h, C24h, C48h) and labeled with Cy3. Each condition consisted of 4 biological replicates. The CVs were computed as indicated in Additional file 1. The median CV is 0.13, 0.10, 0.19, 0.12 and 0.11 for M24h, M48h, M96h, C24h and C48h, respectively.

Figure 4

Improvement of probe reproducibility in microarray v.3. The distributions of probe intensity CV are presented in a series of six boxes (interquartile range) and whiskers plots. Hybridizations were performed with the reference common cRNA pool on microarray v.2 (5 hybridizations, REF_v.2_M24h, REF_v.2_M48h, REF_v.2_M96h, REF_v.2_C24h, REF_v.2_C48h, each with 4 technical replicates per probe) and v.3 (REF_v.3, 12 technical replicates per probe). The cRNAs were labelled with Cy5. The median CV is 0.14, 0.10, 0.16, 0.10, 0.09 and 0.05 for REF_v.2_M24h, REF_v.2_M48h, REF_v.2_M96h, REF_v.2_C24h, REF_v.2_C48h and REF_v.3, respectively.