I am doing some analysis of single cell RNA-seq data from 10xGenomics (protocol). I wanted to infer the library type to I ran RSeQC infer_experiement and got this result:
This is SingleEnd Data Fraction of reads failed to determine: 0.0978 Fraction of reads explained by "++,--": 0.8156 Fraction of reads explained by "+-,-+": 0.0867
Refering to RSeQC documentation, I concluded that we have here SE-like data (10xGenomics data is actually PE but read 1 is used only to keep some barcoding of the cells) and it's stranded. So strandness of reads is concordant with strandness of reference gene.
Now I want to know if we have fr-firstand or fr-secondstrand data. By reading these posts, A: Infer-experiment.py, is strand-specific? and RSeQC infer_experiment.py results interpretation for cuffdiff and featureCounts, I could assume that for SE data, stranded = fr-secondstrand (majority of "++,--") and reversly stranded = fr-firststrand (majority of "+-,-+").
Is it a correct interpretation or did I miss something ?