Tweet
Dear all,
I have recently started working with RNA-seq data. I am looking at protein-coding genes in fat, liver, heart and skeletal muscle tissue in mice. While exploring my data, I noticed that a stretch of the mitochondrial genome seems to be not expressed in any of my samples. It is the stretch that covers genes from COX2 (ENSMUSG00000064354) to ND4L (ENSMUSG00000065947), all on the heavy strand. All protein coding genes including the tRNAs give me zero reads. We know, however, for certain that the gene products should be there, since we made independent protein measurements of genes along the stretch. And mice without all these genes expressed, wouldn't be viable. The reference genome we use for alignment also shows, that there should be genes there.
The other mitochondrial genes show a very high level of expression (above 1000 RPKM).
Then we looked at other datasets, that we have obtained from collaborators and previous experiments. The data was gathered by different researchers and in different labs, using different types of RNA enrichment (ribozero and poly-A enrichment), different species (mouse and human), and at different times. We see the same pattern to a more or less same extend in all the samples!
I was wondering, whether anybody has observed that before or could come up with a reason for what's going on?
Ideas are much appreciated
I have recently started working with RNA-seq data. I am looking at protein-coding genes in fat, liver, heart and skeletal muscle tissue in mice. While exploring my data, I noticed that a stretch of the mitochondrial genome seems to be not expressed in any of my samples. It is the stretch that covers genes from COX2 (ENSMUSG00000064354) to ND4L (ENSMUSG00000065947), all on the heavy strand. All protein coding genes including the tRNAs give me zero reads. We know, however, for certain that the gene products should be there, since we made independent protein measurements of genes along the stretch. And mice without all these genes expressed, wouldn't be viable. The reference genome we use for alignment also shows, that there should be genes there.
The other mitochondrial genes show a very high level of expression (above 1000 RPKM).
Then we looked at other datasets, that we have obtained from collaborators and previous experiments. The data was gathered by different researchers and in different labs, using different types of RNA enrichment (ribozero and poly-A enrichment), different species (mouse and human), and at different times. We see the same pattern to a more or less same extend in all the samples!
I was wondering, whether anybody has observed that before or could come up with a reason for what's going on?
Ideas are much appreciated
Comment