rob-p
commented
1 year ago Ping @k3yavi here. One thing to try @umutcakir would be to see if you get a similar mapping rate outputting a RAD file for processing with alevin-fry. For example if you ran:
salmon alevin -l ISR \
-1 "Single_S1_L001_R1_001.fastq.gz" \
-2 "Single_S1_L001_R2_001.fastq.gz" \
--chromiumV3 \
--sketch \
-i $my_index \
-p 48 \
-o outputfolder \what happens to the mapping rate then? Also, as long as you are careful to pass the files in the same relative order, you can directly pass multiple input read files to the salmon alevin command as a space separated list; you don't have to concatenate them manually.
--Rob
Dear community,
I have 4 lane files for each sample. I want to run Alevin on these samples. Alevin runs without any error, but the mapping rate is too low (about 5%). I tried different k-mers, different versions of transcriptome datasets, and all protocols (--dropseq / --chromium / --chromiumV3), but my mapping rate is still too low.
Let me explain my workflow. These are my samples:
I generated single I1, R1 and R2 files by concatenating fastq.gz files across lanes; for instance, Sample1_S1_L001_R1_001.fastq.gz + Sample1_S1_L002_R1_001.fastq.gz + Sample1_S1_L003_R1_001.fastq.gz + Sample1_S1_L004_R1_001.fastq.gz -> Single_S1_L001_R1_001.fastq.gz
I indexed my transcriptome by salmon, and I used Alevin using the indexed transcriptome. I used the following command:
However, my mapping rate is only ~5%. I used cell ranger on the same data and the same transcriptome; the mapping rate is ~95%. Is there any bug on Alevin? How can I improve the mapping rate?
Thank you!