I am working on some of the microarray data of some of the genes I am intersted in. I wanted to know the expression levels of some of the genes in different cell types. People have deposited data in duplicates or triplicates for every particular cell types and all the microarray experiment was done in the same kind of microarray chip. Now I have to get a mean value for every cell types and compare it with other cell types. Till now I have normalized the data using gcrma package. Now I think I have to normalize the values with any house keeping gene values and proceed furthur. I am not sure how to proceed. I need help. Please guide me through this.
I also have to find the gene coexpression values for all the gene pairs. I calculated the pearson correlation coefficient from the normalized values from all the data sets. Is that okay or I have to calculate PCC only after normalizing the values by any house keeping gene value. Please help me. Thank you
Have a look at this tutorial
https://www.bioconductor.org/help/workflows/RNAseq123/#introduction
If the data is from different studies then you may need to consider possible batch effects. One simpler tool is to use Combat package for batch effects correction, which should perhaps be enough. You still need to look at diagnostics of how similar your samples are after batch correction and normalization. You can find an example for diagnostics at our blogs Laplacebayes.WordPress.com and datacatz.WordPress.com
If its something you understand then I would go down the route of mixed effects models for your analysis and account for batches as random effects in your model.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis