Alternative first exon undetected

[IsabelGam]

Hello,

I recently started to use ASpli to quantify the alternative splicing levels of my gene. My gene has 3 isoforms which only differ from each other by one exon.

• The first one has all exons.
• The second one has an exon spliced.
• The third one is like the first one, but with a different starting exon, which does not overlap with the starting exon from the two other isoforms.

The function binGenome() can detect the exon skipping from the second form. However, it does not detect the splicing event generated by an alternative exon 1.
For now, I modified the GTF file to allow an overlap of the alternative starting exon, so that binGenome() detects it as an alt5'ss event.
I wonder if there is a solution to this problem since I want to quantify alternative splicing levels for other genes, which also have an alternative starting exon (or ending exon).

I'm using R 4.1.0, GenomicFeatures 1.48.3 and ASpli 2.6.0.

I run with the default parameters :

genomeTxDb=makeTxDbFromGFF(file='Gene.gtf', format='gtf')
features <- binGenome( genomeTxDb )
targets <- data.frame(
row.names = name,
bam = file,
factor1 = name)
gbcounts <- gbCounts( features = features,
targets = targets,
minReadLength = 80, maxISize = 10000)

jcounts <- jCounts(counts = gbcounts,
features = features,
minReadLength = 80, threshold = 5)

The BinGenome Log file without my modification of the starting site :

• Number of extracted Genes = 1
• Number of extracted Exon Bins = 18
• Number of extracted intron bins = 19
• Number of extracted trascripts = 3
• Number of extracted junctions = 17
• Number of AS bins (not include external) = 1
• Number of AS bins (include external) = 1
• Classified as:
  ES bins = 1 (100%)
  IR bins = 0 (0%)
  Alt5'ss bins = 0 (0%)
  Alt3'ss bins = 0 (0%)
  Multiple AS bins = 0 (0%)
  classified as:
  ES bins = 0 (NaN%)
  IR bins = 0 (NaN%)
  Alt5'ss bins = 0 (NaN%)
  Alt3'ss bins = 0 (NaN%)

Thanks

[estepi]

Hello Isabel, I see the point. You have a gtf with only 1 gene and there is 1 exon not being terminal, is not considered as AS. It is ok?

Please can you paste the output of this function and tell me which is the bin you want to quantify?

featuresb(features)

thanks for your patience

[IsabelGam]

Hi,

If by "terminal" you mean classified as "external" using the featuresb(features) function, he is classified as such. I have only 1 gene in my gtf, but with 3 transcripts.

The bin I want to quantify PSI values for is E002 -> E003 from the file. (E002 is the alternative first exon).

featuresb.txt

Thanks for you time

[estepi]

Hi Isabel, according our classifier:
E002 E CETP - external external external
is this the exon you want to quantify?

[IsabelGam]

Hi, Yes it is the one.