Dear all, I am totally new for RNA-seq data analysis. Here is my dataset background. There are 3 replicates for Normalized RNA-seq data in 2 conditions. I first want to check How is gene expression profile differences from 2 conditions. So I combine 3 replicates (using mean across 3 samples Q1: Is this correct way to do so?) and check MA plot (it looks fine). However, when I check the MA plots for each sample, I see clearly two clusters of gene expression levels. I am wondering is any expertise can explain me this? Is that because of experiments issue or nature of data? Thanks a lot in advance!
Yes, the plot actually does not look good.
Can you please give more details about your data?
It lack details but based on the figures you presented the normalization procedure was not adequate with respect to the transcript length (or effective length) and library sizes. But then again, we don't know what normalization procedures you have implemented as the statements lack details. You can also generate Loess curve to monitor whether the datapoints were "centered" after normalization. There are methods on how to normalize data points if they are 'misbehaving'. You can also calculate coefficient of correlations between replicates, preferably Spermans.
Hi Lin,
First of all please keep in mind that the MA (ratio intensity) plot is meant to compare two or two group of samples. It concludes how different your samples are in terms of signal intensities (in microarray) or read counts (in RNAseq experiments). If you try to plot sample A versus sample A (same sample) you will notice the data points converge to zero at Y-axis because log (A/A) is zero. Having said that I would recommend you to https://en.wikipedia.org/wiki/MA_plot for basics.
Coming back to your plots it look like that the data is not correct or not normalized correctly (as other people pointed out above). If it is the RNAseq experiment how come the average readcount (x-axes in all your plots) be negative??
I would suggest first normalize or pre-process your data correctly and use any of the bioconductor tools (EdgeR or DESeq2) .
Best
Just a note: If it is the RNAseq experiment how come the average readcount (x-axes in all your plots) be negative?? A: if its log2 scale, its ok to be negative, right ?
For example: http://orange-bioinformatics.readthedocs.io/en/latest/widgets/maplot.html
Easy to understand and visualize MA plot for RNA-seq data: https://reneshbedre.com/blog/ma.html
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis