pooling type: pooled multiple seedlings genotype: Oh43 classification: Non-stiff stalk
Treatment protocol
No treatments
Growth protocol
Maize seeds were grown in a greenhouse with 50% field soil and 50% Metromix with a 16hour light and 8 hour dark treatment
Extracted molecule
total RNA
Extraction protocol
RNAs were isolated using the commercial TRIzol (Invitrogen, Carlsbad, CA; cat# 15596026) from above-ground tissue. RNAs were purified by Lithium Chloride treatment followed by 3M sodium acetate (0.1 vol) and 95% ethanol (2.5 vol) precipitation.
Label
Cy3
Label protocol
Purified RNAs (10ug per sample) were reverse transcribed and labeled according to the array manufacturer protocol (NimbleGen Arrays User’s Guide: Gene Expression Analysis v3.2).
Hybridization protocol
Per sample, ~20μg of Cy3- or Cy5-labeled RNAs were hybridized for 16-20 hours at 42°C using the NimbleGen Hybridization System. Post hybridization, slides were washed (NimbleGen Wash Buffer Kit) and dried for two minutes by centrifugation.
Scan protocol
Slides were immediately scanned using the GenePix 4000B Scanner (Molecular Devices, Sunnyvale, CA) according to the array manufacturer protocol.
Data processing
Array images and data were processed using NimbleScan software. Briefly, images from each slide were separated into 12 subarrays and aligned to a grid to extract signal intensity for each feature on the array. Experimental integrity was verified by evaluation of the signal intensities of the sample tracking control features for each subarray. In addition, metrics reports were produced for each array to report the signal uniformity across the array and the intensity of known empty features, random probes and experimental probes. Signal-to-noise ratios were estimated by dividing the average signal intensity of the experimental gene probes (signal) by the average signal intensity of the random sequence control probes (noise). Only slides with a signal-to-noise ratio ≤ 2 were retained. NimbleScan was used to generate RMA-normalized (Irizarry et al. 2003) gene expression values from the spatially-corrected probe signal intensities on a per probe and per gene basis. Normalized gene expression values across multiple replications (technical or biological) of the same genotype were averaged, when possible. Comparisons of the distributions of signal intensity for the random sequence controls to the experimental gene probes on each array (Figure SM2) were used to determine a reasonable signal threshold for positive expression across all slides. The plots demonstrate that in most cases, the random sequence control log2 signal intensity is ≤ 10 suggesting that signals above that threshold are true signal. Hence, genes with average probe log2 signals of >10 in at least three arrays were retained as expressed (N=19,792 expressed genes). Since genetic polymorphisms could contribute to differences in hybridization between transcripts from some genotypes and probes developed based on the B73 reference sequence we further filtered the probeset based on a previous comparative genomic hybridization dataset developed using many of the same genotypes and the same array platform (Swanson-Wagner et al., 2010). Any probes that exhibit substantially reduced CGH values for at least three genotypes (26,937 probes) were removed resulting in a set of 46,167 probes that detect expression of 18,242 genes with one to four probes per gene. This dataset of N=18,242 expressed genes detected by probes without substantial evidence for polymorphisms was used for all analyses.
Reshaping of the maize transcriptome by domestication
Data table header descriptions
ID_REF
VALUE
Per gene expression values are derived from background-corrected RMA normalized data using NimbleScan (ver2.0) software. Each gene is represented by 1-4 probes that do not show significant comparative genomic hybridization variation.
