First of all, thanks for implementing kallisto | bustools for scRNA-seq analysis. It is super helpful.
I was wondering how does bustools handle multimapping reads. From what I have read, by running kallisto buskallisto generates the bus file and if a read maps to an equivalence class containing transcripts from two different genes in the bustools count step this read will be discarded, right?
I have two questions regarding this.
1) If a read maps to an equivalence class containing several transcripts from the same gene, how would that read be quantified? Would it be discarded?
2) Also, if a read was assigned to an equivalence class mapping 9 transcripts from geneA and 1 transcript from geneB? In that case from what I have read https://github.com/BUStools/bustools/releases , with the -em parameter, it would assigned 0.9 counts to geneA and 0.1 counts to geneB, and with the -m parameter it would assigned 1 count for each gene, right? But with the default options, this read would not be counted.
Hey,
First of all, thanks for implementing kallisto | bustools for scRNA-seq analysis. It is super helpful.
I was wondering how does bustools handle multimapping reads. From what I have read, by running
kallisto bus
kallisto generates the bus file and if a read maps to an equivalence class containing transcripts from two different genes in thebustools count
step this read will be discarded, right?I have two questions regarding this.
1) If a read maps to an equivalence class containing several transcripts from the same gene, how would that read be quantified? Would it be discarded?
2) Also, if a read was assigned to an equivalence class mapping 9 transcripts from geneA and 1 transcript from geneB? In that case from what I have read https://github.com/BUStools/bustools/releases , with the
-em
parameter, it would assigned 0.9 counts to geneA and 0.1 counts to geneB, and with the-m
parameter it would assigned 1 count for each gene, right? But with the default options, this read would not be counted.Are my assumptions right?
Thanks a lot for your help in advance.
Best,
Kike