Bioinformatics Microarray Analysis
At the discretion of our clients, the microarray team can be involved in every step of a microarray project, including analysis. Below are descriptions of standard outputs. For advanced analyses or to request examples or our standard exports, please contact email@example.com
Gene Expression Data
A spreadsheet containing formatted intensity data in log2 scale and with added annotation will be included. The raw chip data .CEL along with .ARR can be loaded into Expression Console Software Expression Console Software or Transcriptome Analysis Console (TAC) Software Transcriptome Analysis Console (TAC) Software (freely downloadable at www.thermofisher.com) to export intensity data, or you can load the raw data into Partek Genomics Suite for advanced analysis (for Wash U researchers, license can be obtained through the Becker Medical Library at the University).
- The off-scanner image files (.TIFF) are processed through the Agilent Feature Extraction software to generate array QC parameters, and to correct abnormal probe performances.
- The output text format of data files contains processed signal and background values of each probe as well as detection calls at the channel used (Cy5 or Cy3) on each probe.
- Data normalization will have to be performed using the quantile or other appropriate method of interest before statistical analysis takes place.
- When sample grouping information are provided with biologic replicates, our analytical pipeline will also perform a standard limma analysis for probe-level differential expression, which is Bayesian moderated t-test with Benjamini-Hochberg multiple test correction algorithm implemented.
SNP And Copy Number
Infinium The off-scanner raw data files (.idat) are processed with the Illumina GenomeStudio software (Genotyping module) for array quality assessment and to generate genotype calls at each locus independently.
- Genotypes are called by comparing sample data with those in the supplied cluster file or customer-generated cluster file. When samples do not perform as expected, experimenters can choose to reprocess these samples (leading to better sample cluster) to confirm or potentially improve results.
- The GenCall score will be calculated as the key quality metric that indicates the reliability of the genotypes called. It is based on information from sample cluster algorithm.
- A GenCall score value can range from: 0 to 1. with a no-call threshold of 0.15 with lnfinium data. This means that genotypes with a GenCall score less than 0.15 are not assigned genotypes because they are considered too far from the cluster centroid to make reliable genotype calls.
- The off-scanner raw data files (.CEL) are processed with the Affymetrix Chromosome Analysis Suite (ChAS) analysis tool for array quality assessment, SNP genotype calls and gene/chromosome copy-number alterations including gain, loss, and loss of heterozygosity (LOH).
- Genotype calls are made using the BRLMM-P algorithm (the Bayesian Robust Linear Model with Mahalanobis distance classifier -only Perfect-match probes).
- Markers on array are individually assigned a copy number call by a Hidden Markov model (HMM), and then the segmentation process was applied to report the gene/chromosome copy-number alterations.
- The default settings for segmentation are as following (1) a minimum segment length should be greater than or equal to 5 markers, (2) small segments of less than 50 markers/200kbp of normal copy number are removed, and the adjacent gain/loss segments are joined.
Infinium HumanMethylation EPIC
The off-scanner raw data files (.DAT) are processed with the Illumina GenomeStudio software for array quality assessment and to generate standard reports. The report will include the methylation b-values. We will also provide the raw data. The GS software is free and you can also load and work on the data from that perspective – review chromosomal coordinates, percent GC, and locations in CpG Islands.
Thermo Fisher Microarray Instruments, Software, and Services
A spreadsheet containing formatted intensity data in log2 scale and with added annotation will be included. The raw chip data .CEL along with .ARR can be loaded into Affymetrix Expression Console (freely downloadable at www.affymetrix.com) to export intensity data, or you can load the raw data into Partek Genomics Suite for advanced analysis (for Wash U researchers, license can be obtained through the Becker Medical Library at the University).