You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi,
I would like to know the detailed rules when splicing variants are determined to be identical in TALON. If there are multiple isoforms and the start and end points of each exon are slightly different, up to how many bp is it considered the same exon?
Also, if the 5' end and 3' end of an isoform are on the same exon as the isofom being compared, can they be combined as the same isoform even if their positions are different? Or is there a limit to how much difference they can be considered the same?
The text was updated successfully, but these errors were encountered:
Hi there,
For splice sites, TALON uses the exact coordinates to define novel isoforms.
For 5'/3' ends, TALON assigns the isoforms the reference 5' or 3' end that's associated with the same intron chain if the ends are within the user-specified distance of the annotated ends (--5p and --3p arguments to the talon_initialize_database command). If a read is found with ends that exceed these boundaries, it will be assigned to a new transcript model. If you're interested in refining your transcriptome post-TALON to have better 5'/3' end calls, I recommend using another tool developed in the lab called Lapa.
Thank you very much for taking the time to answer my question!
Your answer was exactly what I wanted to hear.
However, I did not specifically specify --3p and --5p when I ran talon_initialize_database.
In that case, I thought the default value would be 300bp for 3p and 500bp for 5p.
But when I actually analyzed the data, I found that isoforms that differ significantly more than that are determined to be identical, as shown below.
Why is this?
Hi,
I would like to know the detailed rules when splicing variants are determined to be identical in TALON. If there are multiple isoforms and the start and end points of each exon are slightly different, up to how many bp is it considered the same exon?
Also, if the 5' end and 3' end of an isoform are on the same exon as the isofom being compared, can they be combined as the same isoform even if their positions are different? Or is there a limit to how much difference they can be considered the same?
The text was updated successfully, but these errors were encountered: