Quantification and equivalence classes

The project contains the quantification output of Salmon (including equivalence classes) from simulated RNA-seq data for 12 paired-ended samples (i.e., biological replicates).<br>Reads were first aligned via splice-aware genome aligner STAR, then Salmon was used on the aligned reads to compute transcript-level estimated counts (quant.sf files) and equivalence classes (aux_info/eq_classes.txt files).<br>100 bootstrap replicates are computed from Salmon for each sample.<br><br>Samples belong to two groups: samples 1 to 6 constitute the first group, while samples 7 to 12 represent the second group.In each group 1,000 genes exhibit differential transcript usage (DTU) and further 1,000 genes (partially overlapping with the previous set) shows differential gene expression (DGE) between groups.<br>Genes showing DTU were simulated by randomly permuting the relative abundance of the four most expressed transcripts; for gene with two or three transcripts only, all transcripts relative abundances were permuted.<br>The dataset is used to benchmark DTU methods in the manuscript entitled "BANDITS: Bayesian differential splicing accounting for sample-to-sample variability and mapping uncertainty".<br>The code for simulating the data is available on GitHub at https://github.com/SimoneTiberi/BANDITS_manuscript .<br>