Home → CLC software: Important notifications → Issues affecting only versions of products released prior to June 2017 → Genes with transcripts and with identical names reported as having 0 counts even with mapped reads
3.6. Genes with transcripts and with identical names reported as having 0 counts even with mapped reads
When running the RNA-Seq tool using track based references and
- using the option "Genomes annotated with genes and transcripts", and
- two or more genes had the same name, and
- a transcript could be assigned to each of these genes from the mRNA track
then the value in the "Transcripts annotated" column in the GE track and in the TE track was 0, and all counts for such genes were reported as zero, even when there were reads mapping to them.
Which software and versions are affected?
This issue affects:
- CLC Genomics Workbench 9.0, 8.0 through 8.5.2, and 7.0 through 7.5.5
- Biomedical Genomics Workbench 3.0, and 2.1 through 2.5.2
- CLC Cancer Research Workbench 1.0 through 2.0
- CLC Genomics Server 8.0, 7.0 through 7.5.2, and 6.0 through 6.5.6
Which software versions are fixed?
This issue is fixed in the CLC Genomics Workbench 9.0.1, Biomedical Genomics Workbench 3.0.1 and CLC Genomics Server 8.0.1, released on June 9, 2016.
It has also been addressed in the previous release line: CLC Genomics Workbench 8.5.3, Biomedical Genomics Workbench 2.5.3, and CLC Genomics Server 7.5.3, released June 16, 2016.
The release notes describing the fix for this issue can be found on the Latest Improvements pages for each product. This particular fix is described as follows:
Fixed an issue with the RNA-Seq Analysis tool that could arise when the "Genomes annotated with genes and transcripts" option was chosen: If two or more genes had the same name, and a transcript could be assigned to each from the mRNA track, then the value in the "Transcripts annotated" column in the GE track and in the TE track was 0. Furthermore, all counts for such genes were reported as zero, even when there were reads mapping to them.
How can I check if my RNA-Seq Analysis has been affected by the issue?
To find out if your analysis is affected you:
- Sort the TE track table on the column: "Transcripts annotated"
- Any transcripts that have 0 in this column are affected. All other transcripts are not affected by this bug.
The bug was present at a very late stage in the RNA-Seq algorithm execution, when the calculation results were entered into the table. The underlying calculations were done correctly, but values for the duplicated gene names were excluded. The nonzero values that made it into the table are correct and will not change with the fix.
While transcripts with 0 in the "Transcripts annotated" column (affected transcripts) will often be seen when using the affected software versions under the analysis conditions described above, we do not anticipate that it will be necessary to re-run analyses in most cases. Only a very small number of genes/transcripts would generally be affected and these may not be genes of interest to the analysis. For example, in many cases, genes given the same name code for rRNAs, snoRNA, and other miscellaneous RNAs, rather than mRNAs.
If you do see affected transcripts associated with genes of potential interest, then we recommend re-running the analysis using a version of the software that includes a fix for this problem.