Dear All,
I am very new into this. I am analyzing TCGA data to see the differential expression of a specific gene across samples. I downloaded both rsem.genes.results and rsem.normalized.results. There are many different opinions on this. Which one can I use to do the analysis? There is raw_count and scaled_estimate. From what I read, raw_count can be used as input in edgeR. But scaled_estimate is like transcripts per million and therefore can be used directly with no transformation. And also, I can not use normalized_count to compare across samples, but again I see many papers do a log2 transformation on normalized_count and use that....
I am pretty confused.
Please help
Thanks
I am very new into this. I am analyzing TCGA data to see the differential expression of a specific gene across samples. I downloaded both rsem.genes.results and rsem.normalized.results. There are many different opinions on this. Which one can I use to do the analysis? There is raw_count and scaled_estimate. From what I read, raw_count can be used as input in edgeR. But scaled_estimate is like transcripts per million and therefore can be used directly with no transformation. And also, I can not use normalized_count to compare across samples, but again I see many papers do a log2 transformation on normalized_count and use that....
I am pretty confused.
Please help
Thanks