Salmon provides fast and bias-aware quantification of transcript expression

Abstract

We introduce Salmon, a method for quantifying transcript abundance from RNA-seq reads that is accurate and fast. Salmon is the first transcriptome-wide quantifier to correct for fragment GC content bias, which we demonstrate substantially improves the accuracy of abundance estimates and the reliability of subsequent differential expression analysis. Salmon combines a new dual-phase parallel inference algorithm and feature-rich bias models with an ultra-fast read mapping procedure.

SPLICE-q: a Python tool for genome-wide quantification of splicing efficiency

“…Interestingly, at 0 and 15 minutes, SE scores are already high with a median of 0.71 and 0.75, respectively. This agrees with previous studies showing that splicing is predominantly co-transcriptional in humans and for the most part happens immediately after the transcription of an intron is completed, when the RNA polymerase has proceeded only a few bases into the downstream exon [5, 6, 9–11]. However, the results also illustrate that even 60 minutes after the pulse-labeling of newly synthesized RNA, there is a significantly larger fraction of introns which have not yet been excised from the transcripts than in the steady-state control.…”

Transcriptome and metabolome analysis of crGART, a novel cell model of de novo purine synthesis deficiency: Alterations in CD36 expression and activity

“…HPLC-EC analysis did not indicate accumulation of 5-PRA (data not shown). This is consistent with previous work demonstrating that 5-PRA is highly unstable and breaks down in approximately five seconds under cellular conditions [12,14]. Since it apparently does not accumulate substrate, this suggests that crGART is likely useful as a model of general DNPS deficiency.…”

Benchmarking UMI-based single-cell RNA-seq preprocessing workflows

“…2 B). Workflows based upon pseudoalignment tools required less memory resource and ran faster, which is concordant with previous studies [ 18 , 19 ]. Kalliso bustools and alevin-fry were more than 50 times faster when dealing with 600M reads compared to the second slowest workflow dropSeqPipe .…”