Entering edit mode
6.8 years ago
lhaiyan3
▴
80
Hi,
We have the targeted RNA-seq fastq file for 9 genes. The first 12 bp of any given read in the FASTQ file is a unique molecular index (UMI). We want to generate a matrix with the 9 genes in rows, and the UMIs in columns with counts in the table. Can anyone please give me some suggestions how to analyze the data? Thanks very much for the great suggestions.
best
HY
Probably in your case you might be talking about single cell RNA-Seq data either generated with fluidigm or 10x. Is that so? Then ther are some workflows that helps you do the quality filtering steps and clustering of those data in terms of genes or cells . In your case since I would expect it is only few genes but then the bulk will the starting point of cells at this moment so clustering of cells for the genes will give you the pattern. let me know if its scRNA-seq or not? If its the Ampliseq transcriptome then you can obtain from their proprietory software the read counts as well using CoverageAnalysis, an Ion Torrent Service Plugin. You might have to get an account of the Torrent suite to get this.
RNAseq analysis with Ion AmpliSeq Transcriptome Solution data
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0078644
Thanks. I just know it is not the scRNA-seq data. I will check with them how they get the data.
probably then it will be Ampliseq transcriptome, just confirm it first about the nature of the data.
It is from Qiagen targeted RNA sequencing.
there are some tools listed here . Take a look and see if you can use any of them for your data. But in any case, if you do not have their proprietary software everything should be similar. So run alignment, quantify and get counts, perform PCA and then use for your biological questions the design for downstream analysis.
If your lab has also the QIAgen software or RNA workbench then this link will help else any standard tool.
Great. Thanks very much for your reply. I will try these tools.
If the comments are helpful to put upvotes so that people can use this thread for future reference.
Great. Your suggestion is really helpful for me. I believe other people also will think it helpful. I press the "votes" button.
yes, the like buttons beside the start of my comments. That's how it works. Since the comments so better to do that. If it was an answer then I would have asked you to put it as answer acceptance. Welcome to the community.
Hi, vchris:
Thanks for your great suggestions for me. Sorry I still need your help. Our data is the customerized Qiagen targeted RNAscan panel data. I contacted with the CLC technical support, the CLC biomedical workbench currently did not have the plug in for this kind of analysis. I need to use the command line for the analysis. According to your suggestions, I used the STAR to align the fastq file with the human genome and also use the htseq-count to count the bam file. But the result is not my expected.
Our fastq file is from 9 genes. The first 12 bp of any given read in the FASTQ file is a unique molecular index (UMI). We want to generate a matrix with the 9 genes in rows, and the UMIs in columns with counts in the table. I have the primer sequences and bed file for these 9 genes. Can i do the alignment just for these 9 genes? Do i need to collapse the files according to the UMI? Can you please give me more detailed suggestions about the quantification/gene count for the UMI? We just need the matrix file. I searched a lot but i still did not find the answer. Thanks very much for your great help.
best
HY