Rationale and Background:
CLASS2: A fast and accurate alternative-splicing aware transcript assembler for RNA-seq reads aligned to a genome.
Song L, Sabunciyan S, Florea L. 2016. Nucleic Acids Res. 44(10):e98. [ Medline ]
CLASS2 is a fast and accurate program for transcript assembly of RNA-seq reads aligned to a reference genome. It uses the splice graph model to represent a gene and its splice variants, and a dynamic programming optimization algorithm to score and select a subset of transcripts most likely present in the sample. CLASS2 aims to capture as much alternative splicing variation as possible with very high accuracy.
- A CyVerse account. (Register for a CyVerse account here - user.cyverse.org)
Input File: one SAM or BAM file
-e evidence: the path to the evidence, or annotation, file (GTF format); user can select from the list or use own annotations
-l label: add a prefix and a "_" to the gene and transcript ids in the output GTF file (default: not used)
-F isoform_fraction: do not report transcripts whose abundance level is lower than f*|most expressed transcript| in a gene
--clean: whether to remove the temporary files in -wd (default: yes)
- Optional Output File Name
-o filename: prefix for the output file name
The test data for CLASS2 can be found here: /iplant/home/shared/iplantcollaborative/example_data/class2
- Open CLASS2 in DE
- Choose the input BAM file /iplant/home/shared/iplantcollaborative/example_data/class2/sample.bam
- Optionally select the GTF annotation file /iplant/home/shared/iplantcollaborative/example_data/class2/sample.annot.gtf
- Optionally set other parameters, such as Isoform fraction (-F) , Prefix for transcript names (-l) , Remove depth and splice files (–clean)
- Optionally set the prefix for the output file name
- Launch the analysis
- sample.gtf: Output file
- If selecting 'No' for the '--clean' option, then a folder named "class_tmp" containing alignment, splice junction and read coverage data will be generated.
The following file illustrates the CLASS2 output for the test data, using the provided annotation: