Zygotes were then dechorionated and electroporated as previously described (Christiaen et al., 2009a, c).
Growth protocol
Animals were kept in tanks under constant light in order to avoid spawning. Gametes from several animals were separately collected for in vitro cross-fertilization. Zygotes were then dechorionated and electroporated as previously described.
Extracted molecule
total RNA
Extraction protocol
To sort cells from the B7.5 lineage we used samples electroporated with Mesp>GFP and MyoD(905)>Venus, as described(Christiaen et al., 2008). For each of them, between 1000 and 5000 sorted cells were directly collected in the lysis buffer of the RNAqueous®-Micro Kit (Ambion, Carlsbad, CA) and kept on ice for RNA extraction following the instructions of the manufacturer. RNA samples were stored at -80°C before quality control (QC). QC was performed on Agilent Bioanalyzer 2100 with RNA 6000 Pico Chips and associated kit (Agilent, Santa Clara, CA). Only samples with concentration above 100pg/µL and RNA integrity number above 7 were further considered for microarray hybridization.
Label
biotin
Label protocol
To sort cells from the B7.5 lineage we used samples electroporated with Mesp>GFP and MyoD(905)>Venus, as described(Christiaen et al., 2008). For each of them, between 1000 and 5000 sorted cells were directly collected in the lysis buffer of the RNAqueous®-Micro Kit (Ambion, Carlsbad, CA) and kept on ice for RNA extraction following the instructions of the manufacturer. RNA samples were stored at -80°C before quality control (QC). QC was performed on Agilent Bioanalyzer 2100 with RNA 6000 Pico Chips and associated kit (Agilent, Santa Clara, CA). Only samples with concentration above 100pg/µL and RNA integrity number above 7 were further considered for microarray hybridization.
Hybridization protocol
Single-stranded cDNA were amplified from 500pg of total RNA using the Ovation® Pico WTA System (NuGen, San Carlos, CA), fragmented, and biotin labeled using Encore™ Biotin Module (NuGen) according to the instructions of the manufacturer. QC for concentration and size of the resulting product was performed using a RNA 6000 Nano LabChip and associated kit (Agilent, Santa Clara, CA). Preparation of the target cDNA and microarray hybridization were performed according to NuGen instructions using GeneChip® Hybridization Control Kit and GeneChip® Hybridization, Wash, and Stain Kit (Affymetrix, Santa Clara, CA). Target cDNA were hybridized to the Ciona intestinalis custom-design Affymetrix GeneChip® (Christiaen et al., 2008) for 16 hours at 45°C in a GeneChip® Hybridization Oven (Affymetrix). Washes and staining were performed in a Genechip Fluidics station using the FS450_001 fluidics protocols for expression arrays and microarrays were finally scanned using a GeneArray Scanner, all from Affymetrix.
Scan protocol
ly scanned using a GeneArray Scanner, all from Affymetrix.
Data processing
Raw expression values (.CEL files) for each probe and biological replicate were used, together with previously obtain raw data(Christiaen et al., 2008; Woznica et al., 2012), for normalization and computation of probe set expression estimates using the robust multi-chip analysis (RMA) algorithm (Irizarry et al., 2003). To determine whether we could normalize data from this microarray experiment with a previous one, we performed the following tests: we took the Pearson correlation of expression values for each gene across conditions, which were common to both experiments. To determine the effects of normalizing the experiments together, we normalized each experiment separately, then together, then took the Pearson correlation of each gene's expression value under the separate normalizations with its expression value under the combined normalization. We also identified which genes exhibited significant changes in expression due to changes in normalization, and verified using a training set of validated genes to ensure that normalizing the two data sets together did not adversely affect the signal. LIMMA was used to combine probe-level expression data into estimates of differential expression for probesets, and to compute corresponding p-values (p) using a moderated t-statistic (Smyth, 2004). We used the Benjamini and Hochberg method (BH) to control the false discovery rate. We searched gene models for matches to probe sequences and discarded expression data for those probes mapping to more than one gene model. For Aniseed IDs with multiple probeset matches, we selected a single most representative probeset by using the probeset with the lowest p-value in the COE-COE::WRPW comparison.
Collier/OLF/EBF-dependent transcriptional dynamics control muscle specification from cardiopharyngeal progenitors
Data table header descriptions
ID_REF
VALUE
LIMMA was used to combine probe-level expression data into estimates of differential expression for probesets, and to compute corresponding p-values (p) using a moderated t-statistic (Smyth, 2004).
