could anyone help explain this IGV results?

[sadiexiaoyu]

Dear All,

could anybody help me explain this IGV result (attachment)? they are two RNAseq samples. I have two questions:

1) In the first sample, I am confusing why there are different read coverage for the same gene...(left is high coverage, and right is low coverage)...
2) between the two samples, there are some regions appear in the second sample, but not in the first sample, is that because of the alternative splicing_?

SOS!!!!!

Best,

Sadiexiaoyu

[chadn737]

Are these technical replicates or biological replicates or even different treatments/genotypes? Also, what is the difference in total number of reads for each one.

In RNA-seq, you expect to see differences in the number of reads that map to each gene. That's why you can use it to do differential expression.

I can't tell you whether or not these genes are alternatively spliced. This is not something that can be determined simply by eye-balling it. Statistical tests need to be done and the sources of variation accounted for. Furthermore, without seeing the actual annotations, its rather hard to even know what I am looking at.

[sadiexiaoyu]

Are these technical replicates or biological replicates or even different treatments/genotypes? Also, what is the difference in total number of reads for each one.

In RNA-seq, you expect to see differences in the number of reads that map to each gene. That's why you can use it to do differential expression.

I can't tell you whether or not these genes are alternatively spliced. This is not something that can be determined simply by eye-balling it. Statistical tests need to be done and the sources of variation accounted for. Furthermore, without seeing the actual annotations, its rather hard to even know what I am looking at.

Hi, chadn737,

They are two different treatments of one gene, and this gene is significant expressed in these two treatment samples (DESeq results). But I am confused that why there are different reads coverage in the same treatment of the same gene...

[chadn737]

1) Differences in total number of reads in each sample. If one sample has 50 million reads and the second has 10 million, then you will not have the same number of reads covering a gene.

2) Differences in Expression. You say these are two different treatments. Just because this gene may be expressed under both treatments doesn't mean that it is expressed at the same level. There could be differences in expression due to treatment or even just biological variation. That difference in expression results in differences in read coverage. Its the basis of doing differential expression using RNA-seq.

[sadiexiaoyu]

1) Differences in total number of reads in each sample. If one sample has 50 million reads and the second has 10 million, then you will not have the same number of reads covering a gene.

2) Differences in Expression. You say these are two different treatments. Just because this gene may be expressed under both treatments doesn't mean that it is expressed at the same level. There could be differences in expression due to treatment or even just biological variation. That difference in expression results in differences in read coverage. Its the basis of doing differential expression using RNA-seq.

Hi, Chadn737,

Thank you for your reply. Maybe I did not explain clearly about my question. I am not confused about different reads coverage between the two treated samples, but instead, I am confused about why there are different reads coverage within sample and within one same gene. For example, in the first sample, left side have high coverage, but the right side have very few reads. But all these reads belong to one gene and one sample. That is my question...

Best,
Sadiexiaoyu

[chadn737]

Ahh, that clarifies it a lot. I don't use IGV much and the picture you have posted lacks annotation for the gene, so its hard for me to see what exactly I am looking at. Am I looking at differences in coverage from 5' to 3', 3' to 5' or is there something else unexplained here? Sometimes you see bias towards one of the ends, particularly the 3' end in samples.

[sadiexiaoyu]

Ahh, that clarifies it a lot. I don't use IGV much and the picture you have posted lacks annotation for the gene, so its hard for me to see what exactly I am looking at. Am I looking at differences in coverage from 5' to 3', 3' to 5' or is there something else unexplained here? Sometimes you see bias towards one of the ends, particularly the 3' end in samples.

Hi, chadn737,

it is 5' to 3', and may I ask why there are bias towards one of the ends, especially the 3'end?
Besides, could you help me deal with the second question:between the two samples, there are some regions (left side) appear in the second treatment sample, but disappeared in the first sample, is that because of the alternative splicing?
Thanks!

Best,

Sadiexiaoyu