Substantial equivalence (transcriptomics) experiments
Substantial equivalence (transcriptomics) experiments
In Western Europe, for transgenic crops to be approved for commercial production, proof of their substantial equivalence to conventionally bred crops needs to be provided to the regulatory authorities. One way to establish substantial equivalence between GM crops and conventionally bred crops is to compare the characteristics of the transcription products of the developing seeds and other tissues of the two to detect any unintended transgene effects. The source of this experiment is the "Experimental Guide to Transgenic Technology and Field Characterization of Wheat Crops" [English] H.D. Jones P.R . Hewlett Editors-in-Chief.
Operation method
Substantial equivalence (transcriptomics)
Materials and Instruments
Wheat Move 3.1 Chip Background For more product details, please visit Aladdin Scientific website.
β-mercaptoethanol chloroform isoamyl alcohol SSTE buffer
SDS polyacrylamide gel electrophoresis
We used a wheat cDNA chip with 19846 spots containing 9246 Unigene sequences (http://www . cerealsdb. uk. net/index. htm) for detailed transcriptome studies [3 ]. Duplicate unigene collection arrays were spotted onto a Codelink activation chip (Amersham Biosciences Ltd, UK ).
Microarray hybridization was performed with Alexa fluorescent dyes 555 and 647, the reverse of the fluorescent labeling dyes of the aa- dUTP-cDNA samples. hybridization was performed between two embryos of the transgenic lines (B 102-1-1, B1118-8-4, or B1355-4-2) and the original non-transgenic lines (L 88-31 or Cadenza). Cadenza ) and their original non-transgenic lines ( L 88-31 or Cadenza ) for pairwise comparisons of two endosperm developmental stages (14 and 28 days after anthesis, dpa) and leaves (8 days after germination, dpg) [5]. Each material was sampled three times and tested twice.
The advantage of cDNA microarrays is that they are affordable and allow the user to have full control over the content and design (customized microarrays). In contrast, Affymetrix oligonucleotide chips are more flexible; they are single-dye systems that make the assay simpler, more specific (improved recognition of similar isoforms and members of multiple gene families), and provide more quantitative and easily comparable data.GeneCMp genomic chips, which utilize a set of 25-base oligonucleotides capable of matching any transcribed sequence " probes" that can match any transcribed sequence. In the case of the wheat microarray, there are 11 probes in each set, most of which are aligned with the sequence of the common conserved domains of the assembled Expression Tag (EST). Therefore, in cDNA microarrays, each set of probes corresponds to a number of ESTs rather than to a single EST. in order to recognize different transcripts while fulfilling other conditions, such as a relatively uniform GC content, the probes are designed by program automation. For a target transcript, a perfect matches (PM ) probe and a single base mismatches (MM ) probe were designed, so that there is an estimation and correction for non-specific hybridization. However, there is debate about the true signal value of MM probes, and many widely used analytical methods do not use MM probes (see Section 3.10). Short probes (e.g., 25-mer) are more efficient for sequence-similar transcripts because individual base mismatches are sufficient to destabilize hybridization and the fixed length allows hybridization conditions to be standardized for all probes; in contrast, longer variable-length probes, such as those used in cDNA microarray platforms, will inevitably hybridize with any probe that has sequence similarity to its part, so their true signal integrates several different transcription molecules.
3.2 Plant material and growth conditions
Endosperm and leaves of three hexaploid transgenic bread wheat were used for the transcriptome comparison study. Transgenic wheat lines B102 -1-1 ( L 88-31 background) [ 6, 7] and B1118-8-4 ( Cadenz North background) [ 5] were generated by gene gun co-transformation of two plasmids [ 14]. One plasmid, the p1Axl plasmid [ 13 ], contains the high molecular mass glutenin subunit 1AX1 (Glu-M) gene driven by its own endosperm-specific promoter; the other plasmid contains the selection gene, bar, and the marker gene, uidA, driven by the maize ubiquitin promoter. Transgenic line B1355-4-2 [ 5 ], also from Cadenza, was cotransformed to obtain a "clean" fragment containing only the IAX1 gene and the bar gene coding sequence. A conventional breeding line, L88-18 [ 9], which is sister to L88-31, was also used for transcriptome comparison. A two-by-two comparison of the transcriptomes of the two conventional breeding lines (L 88-31, L 88-18) and the transgenic line (B 102-1-1) was made: B 102-1-1 versus L 88-31, L 88-18 versus L 88-31, and B 102-1-1 versus L 88-18. Comparison of the methods of transformation, i.e., the comparison of "clean" fragments versus the whole plasmid, was also used. fragments versus whole plasmids: B1355-4-2 versus Cadenza, B1118-8-4 versus Cadenza, B1355-4-2 versus B1118-8-4. Detailed information on the relevant gene components carried by the different bread wheats used in this study is given in Table 15 . 1 . 
( 1 ) Plant species in balanced ranks design of potting bowl with 3 biological replicates for each treatment (wheat growth and developmental stages).
( 2 ) Plants were sampled at 14 and 28 days post-flowering endosperm, and 2 plants were planted in each pot. Only 2 tillers per plant were retained. The selected pots and pots (3 biological replicates for this experiment) included the third plant used to take leaves on day 8 after germination.
( 3 ) Spikes were observed daily and marked as soon as the central spikelet was found to be in flower.
( 4 ) Seed endosperm was manually stripped from the glumes under aseptic conditions; a minimum of 24 endosperm was taken from the central part of each of the 2 spikes only for one sample per pot.
( 5 ) Samples were taken at the same time of day to avoid the effects of circadian rhythms.
3.3 SDS-PAGE
The expression of high molecular mass subunit proteins of all wheat lines was detected by total grain protein SDS-PAGE gel electrophoresis ( Fig. 15. 1 ) using a 10% ( m/V ) acrylamide gel and Tris-borate buffer system [ 10 ] .
3.4 RNA Extraction
3.4.1 Wheat endosperm total RNA extraction
The extraction method was adapted from Chang et al [ 11 ].
( 1 ) Grind 2 to 3 g of tissue to a powder in liquid nitrogen using a pre-cooled mortar and pestle (-70°C) (see Note 8).
( 2 ) The ground tissue is rapidly transferred to a centrifuge tube containing 15 ml of extraction buffer (300 μl of β-mercaptoethanol) at room temperature and mixed thoroughly by inversion (see Note 9). 
( 3 ) Extract twice with equal volumes of chloroform:isoamyl alcohol (final volume 15 ml) and separate the liquid phase by centrifugation at 15000 g for 10 min at room temperature.
( 4 ) Add 0.25 times the volume of 10 moI/L LiCl to the supernatant and mix well. precipitate RNA at 4°C overnight, and centrifuge at 4°C and 15000 g for 20 min to obtain RNA.
( 5 ) Suspend the precipitated pellet in 500 ml SSTE.
( 6 ) Extract once with equal volume of chloroform:isoamyl alcohol.
( 7 ) Add twice the volume of ethanol to the supernatant and precipitate at -70℃ for more than 30 min or -20℃ for 2 h.
( 8 ) Centrifuge at 15000 g for 20 min to precipitate RNA.
( 9 ) Wash with 75% ethanol.
( 10 ) Dry the precipitate and dissolve it in water without nucleic acid degrading enzyme.
3.4.2 RNA extraction from seedlings 8 days after germination
The extraction method was adapted from Cheng et al.
( 1 ) A known mass of tissue (approximately 1 g of young leaves) was ground to a powder using a pre-cooled mortar and pestle (-70°C) with liquid nitrogen (see Note 9).
( 2 ) The frozen powder is quickly transferred to a second mortar and pestle with 10-15 ml of homogenization buffer (see Note 10), and grinding is continued until uniform (see Note 11).
( 3 ) The homogenate is transferred to a 50-ml capped round-bottomed spin-cap centrifuge tube and incubated at 60°C for 10 min (final volume of homogenate 5-10 ml).
( 4 ) Cool the tube on ice, add 0.2 times the volume of 5 mol/L potassium acetate at pH 5.5 (see Note 12), mix gently and thoroughly, and let stand on ice for 10-15 min.
( 5 ) Centrifuge at 10000 g for 15 min at 4°C and remove the supernatant into a new centrifuge tube.
( 6 ) Add an equal volume of phenol:chloroform (1 : 1, V/V), mix well, tighten the cap and shake vigorously for 10 min. 10000 g, centrifuge at 21°C for 10 min. Remove the upper aqueous phase into a new centrifuge tube (see Note 13).
( 7 ) Repeat the previous phenol: chloroform extraction and partitioning for the aqueous phase.
( 8 ) Add 0 to 1 times the volume of 3 mol/L pH 5.3 sodium acetate and 2.5 times the volume of ethanol to the aqueous phase. Mix well and incubate at -20°C overnight to improve nucleic acid precipitation efficiency.
( 9 ) Centrifuge at 10,000 g for 30 min at 4°C to precipitate nucleic acids and discard the supernatant. Invert the tube for several minutes.
( 10 ) Dissolve the precipitate with a small amount of nuclease-free water (300-700 μl). Transfer the nucleic acid solution to an Eppendorf tube and add 0.67 times the volume of 10 mol/L lithium chloride to precipitate the RNA. mix well and incubate on ice for 20-30 min (see Note 14).
( 11 ) Centrifuge (15000 g, room temperature, 20 min) to precipitate the RNA and discard the supernatant.
( 12 ) Dissolve the precipitate in the smallest possible volume (200-300 μl) of nuclease-free water, repeat the lithium chloride precipitation, and centrifuge the RNA as in the previous step.
(13) Dissolve the precipitate in 200 μl of nuclease-free water, add 15 μl of 5 mol/L potassium acetate (pH 5.5) and 800 μl of ethanol. Mix well and centrifuge (15,000 g at room temperature for 20 min) to precipitate the RNA (see Note 15).
( 14 ) Discard the supernatant, add 1.0 ml of 80% (V/V) ethanol, and centrifuge to wash the RNA (15000 g, room temperature, 10 min).
( 15 ) Place the centrifuge tube, uncovered, on an ultra-clean bench and dry the precipitate at room temperature for no more than 10 min. dissolve in 100~200 μl of nuclease-free water. Dissolve in 100~200 μl of nuclease-free water. Divide each sample into equal portions to avoid contamination or degradation during repeated freezing and thawing.
3.4.3 Total RNA Samples for Genomic DNA Removal
After RNA extraction, RNA fragments were treated with DNA-free (DNA Enzyme Treatment and Removal Kit, Ambion) according to the instructions. This system is designed to remove DNA contaminants from RNA samples and to remove DNA degradation enzyme I from the processed RNA samples without the need for heat or phenol extraction.
3.4.4 RNA Quantification and Quality Control
RNA concentration, integrity, and quality were examined using a Nanodrop ND 1000 spectrophotometer (Labtech Int, U K ) and an Agilent 2100 Bioanalyzer ( RNA 6000 Nano Assay, Agilent Technologies, PaloAlto, CA, USA) (see Note 16).
3.4.5 Sample Preservation
RNA samples were stored at -20°C for short periods (up to 3 months) and -80°C for long periods.
3.5 cDNA Synthesis
( 1 ) Add 100 μg of DNAase-treated total RNA (no more than 20 μl ), 8 μl of oligo (dT )23-anchored primer, and nuclease-free water to a final volume of 28 μl.
( 2 ) Incubate the starter reaction mixture at 70°C for 10 min and place on ice for 5 min.
( 3 ) Add cDNA synthesis reaction mixture to the starter reaction mixture: 10 μl of 5x first strand buffer, 10 μl of 0. 1 mol/L DTT, 5 μl of 50x aa-dNTP mixture, 2 μl of Superscript Reverse Transcriptase III (200 U/L) and nuclease-free water to a final volume of 50 μl.
( 4 ) Incubate at 42 ℃ for 2~3 h.
( 5 ) Purify the aa-dUTP-dDNA product using a microcentrifuge column (Qiagen) and follow the instructions.
( 6 ) Collect 10 μl of the final eluate as a sample. cDNA product will be used to prepare probes for microarray hybridization and for real-time RT-PCR techniques. The final reaction can be performed without the 50x aa-dNTP mixture (see Note 17).
3.6 cDNA Microarray Labeling
( 1 ) For the cDNA coupling reaction with the burst dye, divide the total cDNA to be labeled with the reverse dye ( see Section 3.5, step 6 ) into two equal portions (5 μl) and place the reactions of the different beam materials in separate tubes.
( 2 ) Add to each tube: 5 μl of aminoallyl-purified cDNA ( see Section 3 . 5, step 3), 3 μl 1 mol/L NaHCO3 labeling buffer, 2 μl ALexa fluorescent dye 555 or 647, and 10 ml nuclease-free water.
( 3 ) Pipette mix and incubate for 1 h at room temperature in the dark.
( 4 ) Remove the dye that is not coupled to the aa-dUTP-cDNA using the mini Elute columns kit (Qiagen) (see Note 18).
3.7 cDNA microarray hybridization
( 1 ) Take 20 μl of mixed labeled cDNA (two Alexa dyes from step 6 of section 3.6) and add to 25 μl of 2x hybridization buffer and 2 ml poly (dA).
( 2 ) Denature the probe at 95°C for 3 min.
( 3 ) The labeled probe is applied to a cover slip and the carrier slip is placed Code Link side down.
( 4 ) Place the microarray hybridization cassette in an oven at 42°C overnight.
( 5 ) Place the chip in a Falcon tube (blue cap) containing Solution A (see Section 2. 6, Step 2) and invert for 15 min at room temperature (see Note 19).
( 6 ) Transfer the chip to a second Falcon tube containing Solution A and invert for 15 min at room temperature.
(7) Transfer the chip to a third Falcon tube containing Solution B (see Section 2.6, Step 3) and invert at room temperature for 15 min (see Note 19). (see Section 2.6, Step 3) and invert for 15 min at room temperature.
( 8 ) Transfer the chip to a fourth Falcon tube containing Solution C (see Section 2.6, Step 3) and invert at room temperature for 15 min. ( 8 ) Transfer the chip to a fourth Falcon tube containing Solution C (see Section 2.6, Step 3 ) and invert for 15 minutes at room temperature.
( 9 ) Place the chip in a dry Falcon tube and dry by centrifugation at 8000 g immediately.
( 10 ) Scan the hybridization chip with an Axon Instruments Gene-Pix 400B Dual Laser Scanner.
3.8 Analysis of cDNA microarray data
The microarrays were subjected to image analysis, and the intensity of the two fluorescent signals at each observation point was examined to assess the differences in transcribed gene expression between pairs of wheat lines. The data were normalized and subjected to statistical analyses appropriate to the model, to describe the experimental design and to test the significance of the differential expression.
3.8.1 Image analysis and normalization
The points on the microarray were scanned into images using GenePix software (Gene Pix version 5, Axon Instruments, USA), and then all points were manually detected to exclude hybridization failures or weak signals. Image analysis gave all pixels for each point, and these data contained the mean and Log2 ratio values of the two fluorescent dye signals (647 and 555 ). These data were used exclusively for differential expression analysis. The data were imported from GenePix into the GeneSpring software package (GeneSpring 6.2 In Silico Genetics, Inc., USA), and then the Log2 ratio values of the signal intensities were normalized. In this study, the comparison between the Log2 ratio value (of differential expression) (M ) and the Log2 product (of brightness) (A ) of a point indicates that the local weighted smooth stroke technique (LOWESS) normalization process can be used to eliminate trends in poor data. In other words, the Log2 ratio values multiplied by (across) the luminance intensity (Log product) should be within a constant range for all points, but the image shows some tendency to change as A is increased. Therefore, the purpose of the normalization process is to eliminate systematic variation in the Log2 ratio data (e.g., due to experimental procedures). The Local Weighted Smooth Stroke Technique (LOWESS) normalization is performed by examining the relationship between M and A values point by point and then adjusting the M value accordingly (adjusted M = MLOWESS- fittedM). Because there were two independent experiments, the data from L 88-31, L 88-18, and B 102-1-1 were processed separately from the data from Cadenza.
3.8.2 Statistical analysis
Referring to the methodology discussed by Kerr [ 15 ], the data were analyzed using the GenStat statistical system (GenStat version 7 , GenStatProcedure Library Release PL15, Lawes Agricultural Trust, Rothamsted Research Harpenden, U K ). Normalized data were analyzed (see [ 5 ] and its supplementary material for further details). Prior to estimating fixed effects for genes, a linear mixed model consistent with Log2 ratios was used to estimate random effects of experimental design (biological and technical replicates). The model estimates the standard error parameters (one parameter per gene) from the overall residuals (noise) of the data through model calculations. The ratio of each parameter to its standard error obeys a t-statistical distribution with residual degrees of freedom, which allows the statistical significance of differential expression starting from 0 ( Log2 range) to be assessed. In our study, significantly differentially expressed genes were screened by expression and copy number (P < 0.05 ), and only those genes with expression differences greater than 1.5-fold and with more than 2 replicates were retained for further analysis.
3.8.3 Verification of differential expression by real-time RT-qPCR
Selected transcripts were subjected to real-time RT-qPCR to validate the microarray expression data, as described in Section 3.13.
3.9 Microarray Data Presentation
For simple substantial equivalence experiments, a scatter plot comparing gene expression between transgenic lines and controls is sufficient. Samples from the literature [ 5 ] that participated in both experiments are labeled in Figure 15. 2. The results are displayed using the GeneSpring software package, plotting the average intensity of each gene pair between each set of comparative wheat lines and highlighting the few genes of interest (statistically significant differential expression). The results are also summarized in Table 15 . 2 for a numerical summary. 
The results show that the transgene did not affect the expression of a significant number of endogenous genes and that the transgenic plants were substantially equivalent to their corresponding non-transgenic controls or parents [ 5 ]. The results also confirmed that the method of transformation (e.g., clean fragments or whole plasmids) had little effect on gene expression patterns. Substantial equivalence experiments emphasize statistical rigor, and for cDNA microarrays, consider issues such as dye preference and microarray spatial variation (see Note 20 for an evaluation of the current cDNA microarray data analysis).
For simple substantial equivalence experiments, a simple scatter plot comparing expression between transgenic and control lines is sufficient; more complex designs may require other displays. Hierarchical clustering, whether on cDNA, oligonucleotide microarrays, or other platforms, is a powerful method for overview of transcriptome data [16]. This method groups samples and/or genes that have genetically linked expression, with different levels of association visualized in a dendrogram. It is usually displayed in a heatmap style: the gene tree is on one side, the sample tree is on the other side, and the colors represent the expression of the genes. If an effect of treatment on expression is detected, samples with different treatments are shown as different branches, with all repetitions appearing on leaves within these branches; in another dimension, differently expressed genes are also clustered together.
Non-hierarchical clustering methods for gene expression profiles, such as K-means, quality thresholds (QT ), and self-organizing mapping are also commonly used. The average expression values of gene clusters can simplify transcriptome data. Co-expression implies common transcriptional regulation and potential functional relationships. Genes within gene clusters are further examined to determine if there are any common known functions (e.g., protein storage, stress response, or defense), or involvement in common pathways. For the vast majority of genes in crops such as wheat, function may only be inferred from sequence similarity. Cluster analysis, display, and annotation tools are available in open resources [ e.g. Bioconductor (http://www .bioconductor .org / ) ] and commercial software packages [ e.g. GeneSpring (AgilentTechnologies, Inc) ].
3.10 Analyzing Wheat GeneChip Backgrounds
The steps to be followed for Affymetrix GeneChip expression analysis are outlined here. Detailed standard procedures can be found in the Affymetrix Gene Expression Analysis Technical Manual (see Note 21).
GeneChip probe chips are manufactured by Affymetrix (see Note 22). Many universities and private companies are now fully equipped with the Affymetrix Gene Chip TM platform and offer a variety of microarray process services (GeneChip probe purchase, cDNA labeling, microarray hybridization, scanning, microarray analysis, etc.).
The Affymetrix Wheat Gene Chip, manufactured by the AffymetrixGene Chip Consortia Program, contains 61,127 probes representing 55,052 transcripts on all 42 chromosomes of the wheat genome. The microarrays were designed based on functional domain data published in GenBank and dbEST ( http://www.affymetrix.com/community/research/consortia .affx ). The Wheat Genome Microarray can be used for gene expression studies in different wheat species: UniGene Build 38, 2004 . 4 . 24 ). The microarray contains ESTs and full-length sequences of all these species up to May 2004 for which probes were designed.
The GeneChip probe chip is manufactured using a combination of photolithography and chemical synthesis techniques. tens to hundreds of thousands of different oligonucleotide probes are placed on the 1.7 cm2 chip, with each probe site (probe chamber) being 20 mm. each target transcriptome is detected by a set of 11 PM and 11 MM probes that are 25 bases in length. These PM and MM probes (probe pairs) are positioned adjacent to each other. Gene expression levels can be calculated from the difference in brightness between the PM and MM probes using Affymetrix software (see Note 22), or from the brightness of the PM probes only (RMA and gcRMA analysis).
3.11 Wheat Gene Chip Expression Analysis
3.11.1 RNA sample preparation
For specific tissues, RNA [ total RNA or purified poly ( RNA species ) ] extraction and purification can be performed using existing procedures (similar to those described in Section 3. 4). For example, TRIZOL - Reagent (Invitrogen, see Note 2 3 ) is recommended for total RNA extraction from wheat flag leaves. When extracts contain large amounts of glycoproteins and polysaccharides, the standard steps of homogenization (step 1 of the TRIZOL RNA Extraction Instructions) and RNA precipitation (step 3 of the TRIZOL RNA Extraction Instructions) need to be slightly modified. For homogenization, an additional step of centrifugation is required for the homogenized product (see Note 24), and for RNA precipitation, aqueous recovery of the precipitated total RNA requires the use of isopropanol and a high-salt precipitation solution (see Note 25). In order to obtain high purity RNA (especially A260/A230. Ratio >1.8), we also recommend that the RNA be washed through an RNeasy column (Qiagen, see Note 26). RNA cleanup is recommended to be performed after the genomic DNA has been removed from the total RNA sample (see Section 3.4). Nucleic acid concentration and quality are determined using a Nanodrop ND 1000 spectrophotometer and an Aglient 2100 Bioanalyzer (RNA 6000 Nano Assay, Agilent Technologies , Palo Alto, CA , USA ), respectively (see Note 16).
3.11.2 cDNA Synthesis and Labeling (see Note 27 )
Double-stranded cDNA is synthesized from total RNA [or purified poly ( A ) RNA], and biotin-labeled cRNA is then transcribed from the cDNA in vitro. The discovery of cRNA fragments prior to hybridization to gene probe microarrays is critical for maximum sensitivity.
3.11.3 Hybridization (see Note 28)
Prepare a hybridization mixture consisting of the cRNA fragments and the probe chip control. Hybridize to the ProbeChip and incubate for 16 h.
3.11.4 Probe-chip elution and staining
( 1 ) Fluidic station setup: The Fluidic station is used for chip elution and staining. It is operated through the GeneChip Operating System (GCOS)/Microarray Suite on a compatible PC workstation. Steps include setting up and starting the Fluid Station (see Note 29 and Note 30).
( 2 ) Washing and Staining of Microarrays (see Note 31) : After 16 h of hybridization, remove the hybridization solution from the microarray (see Note 28), and then fully immerse the microarray in the appropriate volume of the recommended wash solution (see Note 31).
3.11.5 Chip scanning (see Note 31)
Once the scanning is complete, each complete chip image is saved in a file with a .dat extension and the name of the experiment.GCOS collects and analyzes the chip images and experimental data: defining the probe units and calculating the light intensity of each unit ( see Note 22). Due to higher quality control during manufacturing, many of the problems of cDNA microarray profiling, such as spatial variability, are not considered for Affymerix chips (Section 3.8). An output file (cel file) containing the signal values for each PM and MM probe was generated.
3.12 Wheat GeneChip Data Analysis
Affymetrix microarrays are widely used in many species of organisms, and considerable effort has been invested in developing and testing different methods to analyze the signals in order to find the best method for gene expression detection.The method developed by Affymetrix is to estimate the expression as the average difference between the PM and MM of a set of probes (MAS5 ). However, the validity of the MM signal values is questionable, and there are a number of alternative methods that appear to in fact exceed MAS5. The RMA (robustmultichipaverage) algorithm takes all the PM data (e.g., all the cel files) for an experiment and not only normalizes the median expression data for each microarray, but also applies the same variance to each microarray expression data [ 17]. 17]. gcRNA algorithm is a variant of RMA that takes into account the weight of each probe's GC composition on the signal contribution [18]. Compared to other methods, it handles Affymetrix microarray data well for labeling [ 19 ] detection and real-time reverse transcription PCR quantification [ 20 ]. gcRNA algorithms can be used with the open-source Bioconductor software package (http :// w w w .bioconductor .org) or commercial software such as GeneSpring7 (Agilent Technologies, Inc).
Once an expression assay has been selected (e.g. MAS5, RMA, gcRNA, or other), the subsequent analysis is the same, whereas non-normalized data (e.g. MAS5) must first be normalized (e.g. by dividing by the median expression value of each microarray). It is recommended to first screen the probe set, retaining all probes with absolute expression values above a threshold (at least for one sample) (which can be judged from the signal of the non-wheat control on the microarray). These probes are further screened for those with expression differences above the threshold for any pair of samples; typically a 1.4-fold change is considered the minimum that can be detected. Substantial equivalence experimental designs typically have more than 2 genotypes with at least 3 biological replicates. In order to detect genes that are statistically significantly differentially expressed between genotypes, an analysis of variance (ANOVA) is performed on the logarithm of the expression values for each probe set, which are usually log-normally distributed. Assuming that the number of probes is still very large even after screening, a multiple testing correction can be used. the Benjamini-Hochberg false positive rate (FDR) correction [ 21 ] is a good choice. The classic P < 0.05 ANOVA and Benjamini-Hochberg multiple test correction for many probes can be used, and it is purely a matter of chance that about 5% of the genes will pass. However, if few or no genes pass this correction, multiple testing can be canceled and the ANOVA results are the result of a false positive rate. For example, if 1,000 probes are tested at P < 0.05, and 50 pass without the multiple testing correction, this is just not more than expected luck. The substantial equivalence criterion for the list of significant genes is the same as for cDNA microarray experiments [ 5 ].
3.13 Real-Time RT-PCR Validation of Transcriptome Data
There are two commonly used methods for quantitative gene (amplicon) detection: gene-specific fluorescent probes (e.g., TaqMan chemistry) or specific double-stranded DNA binding reagents (SYBR green chemistry ) [22]. We chose SYBR green chemistry to verify the expression of DEGs found by cDNA microarrays by real-time RT-PCR. Specific primers were designed for different genes (see Note 32).
3.13.1 Preparation of the real-time RT-PCR reaction
( 1 ) PCR was performed with a visualized 96-well plate on an ABIPRISMA 7500 Sequence Detection System instrument ( Applied Biosystems, Foster City, CA, USA ).
( 2 ) Total RNA ( 2 μg of deoxyribonuclease-treated RNA from section 3.4) was reverse transcribed with reverse transcriptase and buffer (Superscript EIRT, Invitrogen) according to the instruction manual.
(3 ) The PCR reaction was performed with a 25 μl system: 100 ng of cDNA, 12.5 μl of 2X Platinum qPCR Super Mix-UDG with SYBR Green Fluorescent Dye (Invitrogen), 0.5 μl of ROX Reference Dye (Invitrogen), and a pair of specific primers (200 ng of each primer) (see Note 33). ).
( 4 ) All PCR reactions were controlled with the following standard temperatures: 50°C for 2 min, 95°C for 2 min, 40 cycles: 95°C for 15s and 60°C for 1 min ( see Note 34 ).
3.13.2 Real-time PCR data extraction and analysis
( 1 ) Raw data extraction. The cycling threshold (Ct ) was collected for each sample [23]. In order to compare the Ct values of d