Hi All,
I have aligned my fastq files using hg19 and hg38 genomic assembly using HISAT and i have performed differential gene expression analysis based on both the genomic assembly using cufflink, count and stringtie based protocol (table.1) . In this analysis i have obtained lesser number of differentially genes are common between both genomic assemblies. Have anybody face similar kind of results? Which method is reliable in this case? please help me
for example:
table.1 Number of differentially expressed genes obtained using hg19 and hg38 genomic assembly.
| HG19 | HG38 | common |
cufflink | 302 | 165 | 95 |
count | 80 | 73 | 28 |
stringtie| 128 | 148 | 21 |
•
written
by
akilabioinfo •
You're likely observing a few different things. Firstly, hg19 has a number of problematic regions that cause aberrant alignments. These were largely fixed in hg38, which is one of the reasons people have transitioned to it. The other thing you're observing is that using different numbers and sizes of genes will result in different counts and thus different filtering and results from multiple testing adjustments.
For what it's worth, there is never a reason to use cufflinks any more. The authors don't even recommend using it. There's rarely a benefit to using stringTie in human data.
There's rarely a benefit to using stringTie in human data.
I did not knew that! Can you please elaborate?
The point of stringTie is to assemble new transcripts that aren't annotated. There aren't many of those left for things like mouse and human.
Isn't StringTie a reference guided assembler and it provides both kinds of transcripts?
Yes, but if you just want to quantify known transcripts then salmon or kallisto will do that MUCH faster.
There's rarely a benefit to using stringTie in human data. Is there is any reference for stringtie performance on human data?
You can search pubmed for one of the numerous comparisons of differential expression programs. StringTie is useful when you have a poorly annotated organism or when you're interested in finding novel isoforms. Other than that there's never been a compelling reason to use it.
Log in to answer this question.
