These results are the part of two-channel microarray experiments-
Each gene was in triplicate on the array and each sample was hybridized with a flip-dye replicate in order to account for labeling bias.
Genes were identified by SAM (multi-class) analysis of whole genome transcription at 0h, 3h, 6h, 9h, 12h. At each time point logFC was calculated by comparison of parallel grown control vs bacteria infected epithelial cell. The attached sheet shows the heatmap and data-values.
Is it possible to convert this figure from 10 lane to 5 lane, means if I average the logFC for two columns like for first two block of 0hr to [(0.64+ (-1.15)]/2 = -0.255. Is it will be correct to take average of lanes?
once you have a dataframe containing rows as gene names
coloumns as your conditions
just draw a heatmap in R
library(gplots)
library(RColorBrewer)
library(genefilter)
percentage
I thought you did need to average both two columns and three rows before heatmap.
If you want to compare them with single-channel data, you can also use our equations (http://www.ebiomedicine.com/article/S2352-3964%2814%2900014-0/pdf) to convert two-channel data into single-channel data.
You can use Gene-E (www.broadinstitute.org/cancer/software/GENE-E/) to average your columns and generate a new heatmap in a couple of clicks. Use the collapse tool to merge columns by means, average, etc.. and use the clinical annotations to select the columns to collapse (see images bellow). You can use InSilico DB (https://insilicodb.com) to export your excel files to Gene-E.
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology