Thanks a lot! When using tximport, found that txi$counts and txi$abundance has a big difference, thought txi$abundance should be used as the TPM value, is this right? Thank you!
tximport(file, type="kallisto", tx2gene=t2g,ignoreTxVersion=T, countsFromAbundance = "lengthScaledTPM")
I guess so, but you could simply calculate gene level TPMs manually with the counts and length information that tximport gives you, using this simple formula: Raw counts to TPM in R
There is a formula to TPM that will give you the exact relationship:
What the FPKM? A review of RNA-Seq expression units
The TPM reflects the abundance of a given transcript relative to all other transcripts. Thus the TMP value is determined not just by the count for that row but by all other counts over all other transcripts. Hence you can't quite estimate what a TMP would be from a single count alone.
Log in to answer this question.
