diff --git a/README.md b/README.md index bc6a49f7812e4ba49ac3386e8710a9beed0bc067..04fdab6881eb4a6f64362139757909f5dc3df49d 100644 --- a/README.md +++ b/README.md @@ -1,7 +1,7 @@ jSplice 1.0.3 ------------- - -jSplice is a fast method to detect differential alternative splicing events from RNA-seq experiments. (see publication for details). +jSplice is a fast method to detect differential alternative splicing events from RNA-seq experiments. The method relies on junction reads (mapped reads that span an exon-exon junction) to uncover differentially expressed alternative splicing modules (ASMs). AMSs were introduced by Hu et al. (Nucleic Acids Research 2012) and have the advantage to provide a broad definition for alternative splicing events while avoiding the whole transcript reconstruction problem. +The method was compared to DEXSeq, MISO, MATS, DiffSplice, and SplicingCompass and displayed a better accuracy while providing a tremendous speedup (see publication for details). jSplice is freely available under the GPL license at http://www.mhs.biol.ethz.ch/research/krek/jsplice @@ -30,7 +30,6 @@ If you don't have administrator rights, then append "--user" to the command. Als Running jSplice --------------- - jsplice [-h] [-t KEYWORD] [-e] [-b] [-n N] [-s] [-S] [-j] [-x NBEXPS] [-c COUNT] [-r RDIFF] [-k RPKM] [-i INCL] [-p PVAL] [–a ANNOTATION] –d DESIGN –o OUTDIR Upon the first run, jSplice saves the results into an object file (“OUTDIR/jSplice.datâ€). If one wants to re-run jSplice with different parameters or thresholds, the object is loaded to lower computational costs. @@ -40,7 +39,6 @@ Note: if the jSplice folder is in your PYTHONPATH then you can call it as a modu Required parameters ------------------- - -o, --outdir: Output directory. Any file already present in this directory will be overwritten. Ideally, the directory should identify your experiment. After execution, the directory will contain a jSplice object, a BED file of the trimmed genome, and all files generated by coverageBed (if any). Note that the directory will be used by runJSplice.py later on. -d, --design: Path to the experimental design file. The text file should contain one line per experiment and four columns: experiment, condition, junction file, and BAM file. If –b is set, then a BED file is expected in the fourth column instead of the BAM file. If –j is set, then the fourth column can be skipped. @@ -50,7 +48,6 @@ This parameter is only required for the first run. Optional parameters - controlling the thresholds ------------------------------------------------ - -c, --count: Read count threshold. Any junction or exon that does not reach the threshold in at least one condition for all experiment is discarded. Default is 20. -r, --relfc: Relative fold-change threshold. The ratio difference of two element of an ASM is computed as the difference between the fold-change of condition A to B. The pair of elements that maximizes the fold-change difference (or relative fold-change) across all experiments is selected as representative of the ASM. The average log2 ratio difference of the representative pair is used to sort ASMs. If its value does not meet the threshold then the ASM is discarded. Note that a fold-change ratio is expected, not its log2 transform. @@ -68,7 +65,6 @@ Optional parameters - controlling the thresholds Optional parameters - others ---------------------------- - -h, --help: Display the help menu. -j, --jxnonly: Flag to trigger the junction-only mode. If present, exon will not be considered in the analysis. This allows bypassing the execution of coverageBed which can be time-consuming. Note however, that without exons, retained introns cannot be identified. If set, the experimental design file does not need to contain any BAM or BED files. Three columns are sufficient: experiment, condition, and junction file.