diff --git a/README.md b/README.md index 8c8c25e..fc6e9d5 100644 --- a/README.md +++ b/README.md @@ -24,54 +24,21 @@ your sample preparation protocols etc. This is where this program comes into play. ## Running -Download the **jar** from [Releases](https://github.com/chhh/deltamass/releases/) section. - -##### *Command line* -Run `java -jar okde-x.x.jar` to see the help. -There are quite a few options there and it might seem overwhelming at -first, so you might want to try the GUI version first. +Download the **jar** file from the [Releases](https://github.com/chhh/deltamass/releases) section. ##### *GUI* -Run `java -jar okde-x.x.jar --gui` to launch the GUI. You might want to +- On Windows you can just start the `.exe` file from the [Downloads](https://github.com/chhh/deltamass/releases). +- Run `java -jar deltamass-x.x.jar gui` to launch the GUI. You might want to include a JVM switch `-Xmx2G` (change the number to the amount of gigabytes of memory you'll allow the process to consume), so the command becomes: -`java -Xmx2G -jar okde-x.x.jar --gui` +`java -jar -Xmx2G deltamass-x.x.jar gui` -## Input -The program takes **.pep.xml** files as input. It can accept whole directories -and traverse them recursively as well. See `-i` command line option. -If your _.pep.xml_ files are named differently, use `-ip` or `--in_pattern` to -change the regular expression used for matching file paths. -**Note**: the regular expression is matched against the whole absolute path for -each file, so you almost always want to start the regex with `.*`. Use double -backslash to escape characters, e.g. `\\.` to match a literal dot. - -## Most important settings - * `-g` or `--gui` - run in GUI mode. - * `-i` or `--in` input paths, can be a comma separated list (don't use spaces). - If one of the files is a directory, then it will be traversed recursively and - all files within will be matched against a regular expression provided in - the `-ip` option (which defaults to _*.\\\\.pep\\\\.xml_). - * `-ip` or `--in_pattern` - a regular expression for matching files found in - paths of `-i`. The default is `.*\\.pep\\.xml`. - * `-k` or `--kernel` - the kernel to be used in KDE. Default is *EPANECHNIKOV*, - which is very fast, but the shape of the peaks might suffer a bit depending on - your sample size. If you want smoother curves try *GAUSS_FAST* instead. - * `-b` or `--bandwidth` - list of bandwidths to be used for generating KDEs. - You can think of bandwidth as of the area of influence of each input data - point. E.g. `-b 0.1,0.001,0.0001` will plot three KDEs in the same plot using - these three bandwidths. - * `-hd` or `--h_dynamic` - will automatically estimate the best bandwidth - at each nominal mass separately. Thus, the are under the whole KDE might not - be unity anymore, but the plot might be a lot nicer. It also takes more time, - as the estimation requires looping through data. - * `-p` or `--peaks` - if peaks should be detected. If KDE is being plotted - (using `--plot` which is on by default), then peaks will be overlaid over the - plot. **Note**: peak picking is very parameter dependent. In general it will - depend on the number of data points (affected by the total data range and the - sampling rate `-ms`) and most of all on the denoising, specified in - `--denoise` and `denoisingParams` options. See help for corresponding options - and just try different parameters for yourself. +##### *Command line* +Run `java -jar deltamass-x.x.jar` to see the help, it will print available *commands*. +To print help/usage for a specific command add `command -h`, e.g. +`java -jar deltamass-x.x.jar peaks -h` +There are quite a few options there and it might seem overwhelming at +first, so you might want to try the GUI version first. ## Plot navigation * Left Mouse Button = __LMB__, Right Mouse Button = __RMB__. @@ -90,57 +57,4 @@ backslash to escape characters, e.g. `\\.` to match a literal dot. results that fall into this range. * `RMB click` - options for the plot (__Preferences__), you can also save the plot from here (prefer saving to _PNG_) and change how zooming works. - - -## Starting tips - * First do a preliminary run with low mass step (e.g. `-ms 0.01`) and low - bandwidth (e.g. `-b 0.01` or `List of bandwidths` in GUI) to get a bird's eye - view of your data. - * Typically you'd want to look at the data using `-ms 0.0001 -b 0.001` to - see individual peaks better. - * The peak at zero is always the largest one, and the situation to the left - (negative mass shifts) and right (positive) of it are quite different, so it's - better to view them separately and avoid the zero peak at all. - * Most of your data points will be concentrated in the zero peak, - calculating KDE for it is relatively expensive and there is little merit to it. - You can easily avoid the peak at zero at all by adding e.g. `-ml 0.75` option - (`Mass Lo` in GUI), which will only plot mass differences above mass of 0.75. - * Similarly for negative massses, use `-mh -0.75` (`Mass Hi` in GUI) to plot - everything from negative infinity up to -0.75. - * Use `-ml` and `-mh` options simultaneously to limit the view and increase - plot's responsiveness. - * You can output the detected peaks to a file using `-o` option and skip - plotting completely by providing `--plot false`, however you'll likely - want to do the other thing around most of the time. - -## Filtering data -Input data can be filtered according to the scores in _pep.xml_ files. All then -input files must have this corr in order for it to work. E.g. if you want -to filter based on _peptideprophet_ probability, all the files must have between -processed with PeptideProphet. - -* Filters are provided using `-s` or `--corr` option. The GUI provides a -builder for the filter strings. -* The names of the scores are exactly as they appear _pep.xml_ (e.g. `expect`, -`hyperscore`, `interprophet`, etc). The GUI has a dropdown menu for the most -common ones. -* Filters can be combined using commas. -* Supported operators are `>`, `>=`, `==`, `<=`, `<`. - -Example: `-s "expect<=0.001,peptideprophet>0.95"` - -## Caching -The input data are _.pep.xml_ files, which are typically large, the included -parser for them is not very fast, so if you want to play around with parameters -it makes a lot of sense to cache the parsing results. The option to cache -.pep.xml contents is on by default in both GUI and command line versions. -##### *Disabling* -The cache files are stored nearby your original files. If for some reason you -don't want those extra files to be created use `--cache false` option or untick -`Use cache` checkbox in GUI. -##### *Deleting* -If you've already run the program with cache option and now want to get rid of -those extra files, run the program with `-x` or `--delete_cache` switch, This -won't trigger any processing and will only look for cache files in the input -paths (`-i`) and delete them. The GUI has a `Delete cache` checkbox in *Files & -Cache* section. + \ No newline at end of file