how to measure gene expression level using third sequencing long reads (for example, PacBio long reads)

Question

[bioxc]

When I read some papers which have used third sequencing technology, e.g. PacBio iso-seq, I didn't find any paper which use the long reads to measure gene expression level. They did RNA-seq by third sequencing, but measured gene expression level by second sequencing short reads. Can someone tell me why or give an example using third sequencing long reads to measure the gene expression level? Is it possible to use the third sequencing long reads to estimate gene expression level?

Answer 1

[GouthamAtla]

As far as I know, PacBio sequencing is not quantitative. The main aim of ISO-Seq is to to sequence end to end transcripts, to identify the novel transcripts and to get the full length transcript structure. Its qualitative but not quantitative. For quantitative RNA-Seq, you need to use illumina "like" sequencing platforms.

[bioxc]

Thanks, could you explain to me more about the non-quantification of PacBio sequencing?

[prasundutta87]

Good question..As far as I understand, pacbio reads are long reads that are better than short reads (like illumina) as they produce less gaps. PacBio reads are mainly used for genome assembly purposes and do not need or have much depth which is a necessity for gene expression/variant calling studies that depend on high depth of short reads..Correct me if I am wrong..

[bioxc]

But what is the purpose of the ISO-seq of PacBio? It is for transcriptomics.

[GouthamAtla]

I update the answer.

[bioxc]

Thanks! Why can't ISO-seq be quantitative? The main difference between ISO-seq and illumina RNA-seq is the reads length, so the sequencing results should be the number of reads of each isoform, if so, why don't we normalize them by the library total reads to estimate the gene expression level, like RPM (reads per million)?