Closed HaiderMDev closed 1 year ago
Hi,
in brief, JACUSA2 models sequencing data with the Dirichlet-Multinomial distribution and performs a likelihood-ratio test.
Please check manual.pdf page 12 for details.
When only one condition I is provided, call-1 will create an in silico condition by using available reference information to replace non-reference base calls and creating synthetic base call vectors and applying the likelihood ratio test defined above. The output will contain only data for the non in silico condition.
Hello,
Thank you for your response. So, just to be accurate, when you use the call 2 function then Jacusa2 is modeling the sequencing count data using the Dirichlet-multinomial distribution and using the likelihood ratio chi-squared test for analyzing statistical significance of the difference between the two samples. Therefore, you are directly comparing two samples when using call 2, and not using any information from a reference genome?
When using call 1, you can either compare the sample to an in silico condition or to the reference genome?
Thank you for your answer in advance. I really appreciate it.
Hello,
I have a rather simple questions. When I am calling call 2 for Jacusa2, is it doing call 1 for each of the BAM files before the comparative analysis? Thank you.