Tweet
Hi all,
I am using Cuffdiff and DESeq to determine deferentially expressed genes in s. pombe. I get pretty different outputs with some overlap ie. cuffdiff 13 genes and DESeq 228 and between the two methods 2 are the same. I use the same annotation files for both and as close to as possible the same model parameters. At this point in the analyses i am ok with the differences i contribute them to different methods count vs FPKM and that Cuffdiff has extra stats to deal with isoforms. However my bosses aren't satisfied with the explanation (i am the only computer person fighting with a wet lab group).
Are there any other known biases between the two methods that would help to explain this to my bosses or is there any thing i can do with the data to help visualize the biases? Any help in better explaining this to my bosses would be greatly appreciated as we are on a standstill with this project until i can convince them the difference is ok.
Thanks!
I am using Cuffdiff and DESeq to determine deferentially expressed genes in s. pombe. I get pretty different outputs with some overlap ie. cuffdiff 13 genes and DESeq 228 and between the two methods 2 are the same. I use the same annotation files for both and as close to as possible the same model parameters. At this point in the analyses i am ok with the differences i contribute them to different methods count vs FPKM and that Cuffdiff has extra stats to deal with isoforms. However my bosses aren't satisfied with the explanation (i am the only computer person fighting with a wet lab group).
Are there any other known biases between the two methods that would help to explain this to my bosses or is there any thing i can do with the data to help visualize the biases? Any help in better explaining this to my bosses would be greatly appreciated as we are on a standstill with this project until i can convince them the difference is ok.
Thanks!