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
I found that the counts' values showed in sashimi_plot are different from the counts' value in SE.MATS.JC.txt. I was wondering why this happens?
For example:
In SE.MATS.JC.txt
2 "TMEM181" "TMEM181" chr6 + 159028275 159028375 159026298 159026379 159029364 159029495
2 240 2 100 50 NA NA 0.984 NA
The IJC_SAMPLE_1 value is 240, but in sashimi_plot
the counts value= 98 +114, which is not equal to 240.
Thanks in advance!
Best,
Kun
The text was updated successfully, but these errors were encountered:
The counts are not expected to be exactly the same because rmats2sashimiplot uses MISO to do the counting and generate the plot. The count information from SE.MATS.JC.txt is not directly used. One possible explanation is that rmats may have filtered out some of the reads (maybe for not matching the read length or having extra cigar operations). There's a similar issue about the count differences with more details: #33
Hi,
I found that the counts' values showed in sashimi_plot are different from the counts' value in SE.MATS.JC.txt. I was wondering why this happens?
For example:
In SE.MATS.JC.txt
The IJC_SAMPLE_1 value is 240, but in sashimi_plot

the counts value= 98 +114, which is not equal to 240.
Thanks in advance!
Best,
Kun
The text was updated successfully, but these errors were encountered: