Hi all- I have coverage files from MNase-seq experiments in bigWig format which have been normalized to show nucleosome occupancy at each position in the genome. I have been very successful in using these coverage files in deepTools, along with bed files containing regions of interest (including TSS) to show occupancy at these regions and in flanking sequences. Now, I would like to add an extra layer to this analysis. Rather than simply aligning around the TSS, I would now like to sort the heatmaps by gene expression. So, I have coverage files, and I have access to some RNA-seq data in BAM format. What are the steps I would need to take in order to sort by gene expression? I would prefer to continue using deepTools if possible. I know that R is often used to generate heatmaps as well, but the R language seems far more complicated than the command line options in deepTools. Keep in mind that I have never performed RNA-seq analysis, and that I am relatively new to bioinformatics in general. Thanks!
Question: Sorting heatmaps by gene expression
3.9 years ago by
mmmmcandrew • 100
mmmmcandrew • 100 wrote:
ADD COMMENT • link •
3.9 years ago by
Devon Ryan ♦ 97k
Devon Ryan ♦ 97k wrote:
You have two methods of doing this within deepTools:
- Presort your BED file according to expression.
- Use that with computeMatrix, which now defaults to
- If you use
plotHeatmapthen the results will still be sorted according to expression.
- Run computeMatrix as normal on the MNase samples, ensure that
--sortRegions keepis used.
- Run computeMatrix with the RNAseq files, likely using the --metagene option. Again, use
- Merge the output of 1 and 2 with
- When you use plotHeatmap, use the
--sortUsingSamplesoption and specify the last sample (it's numeric, so if you have 6 MNase samples and one RNAseq sample then specify 7).
ADD COMMENT • link
Please log in to add an answer.
Powered by Biostar version 2.3.0
Traffic: 2038 users visited in the last hour