Module: rje_disorder Description: Disorder Prediction Module Version: 1.6.0 Last Edit: 21/08/20

Copyright © 2006 Richard J. Edwards - See source code for GNU License Notice

Imported modules: rje

See SLiMSuite Blog for further documentation. See rje for general commands.

Function

This module currently has limited function and no standalone capability, though this may be added with time. It is designed for use with other modules. The disorder Class can be given a sequence and will run the appropriate disorder prediction software and store disorder prediction results for use in other programs. The sequence will have any gaps removed.

Currently six disorder prediction methods are implemented:

• IUPred2 : From V1.5.0, IUPred2 API calls can be made to https://iupred2a.elte.hu/iupred2a/. The IUPred method
needs to be given as iumethod=X, where X is short, long, anchor, or redox. This can also be triggered by setting disorder=X to iushort2 or iulong2, iuanchor2 (or simply anchor2), or iuredox2.
• IUPred: Dosztanyi Z, Csizmok V, Tompa P & Simon I (2005). J. Mol. Biol. 347, 827-839. This has to be installed
locally. It is available on request from the IUPred website and any use of results should cite the method. (See http://iupred.enzim.hu/index.html for more details.) IUPred returns a value for each residue, which by default, is determined to be disordered if > 0.5. The IUPred method needs to be given as iumethod=short or iumethod=long. This can also be triggered by setting disorder=iushort or disorder=iulong.
• FoldIndex: This is run directly from the website (http://bioportal.weizmann.ac.il/fldbin/findex) and more simply
returns a list of disordered regions. You must have a live web connection and curl on your system to use this method!
• ANCHOR: Meszaros B, Simon I & Dosztanyi Z (2009). PLoS Comput Biol 5(5): e1000376. This has to be installed
locally. It is available on request from the ANCHOR website and any use of results should cite the method. (See http://anchor.enzim.hu/ for more details.) ANCHOR returns a probability value for each residue, which by default, is determined to be disordered if > 0.5.
• Parse: Parsed disorder from protein sequence name, e.g. DisProt download.
#X-Y = disordered region; &X-Y = ordered region [0.0]

• IUScoreDir: From V1.2.0, pre-calculated disorder scores can be loaded from a file (see below). To activate this mode,
the disorder=X setting should match the <DISORDER> part of the filename (below). If this is one of the other recognised disorder predictors above, missing files will be generated if possible. Otherwise, files must be present for disorder prediction to occur.

For IUPred, the individual residue results are stored in Disorder.list['ResidueDisorder']. For all methods, the disordered regions are stored in Disorder.list['RegionDisorder'] as (start,stop) tuples.

IUScoreDir

V1.2.0 introduced the optional use of IUScoreDir/ to save and/or (re)load lists of disorder scores. These are files named <ACC>.<DISORDER>.txt where <ACC> is the accession number of the protein. If md5acc=T then an md5 hash of the sequence is used instead: hashlib.md5(<SEQUENCE>).hexdigest().

Commandline

General Options

disorder=X : Disorder method to use (iupred2/iupred/foldindex/anchor/anchor2/parse) [iushort2]
strict=T/F : Whether to exit with error if disorder method not found [False]
iucut=X : Cut-off for score-based method (e.g. IUPred/ANCHOR) results [0.2]
iumethod=X : IUPred method to use (long/short/redox/anchor) [short]
sequence=X : Sequence to predict disorder for (autorun) []
name=X : Name of sequence to predict disorder for []
minregion=INT : Minimum length of an ordered/disordered region [0]
minorder=INT : Minimum length of an ordered region; over-rides minregion if >-1 [-1]
smoothing=X : Smoothing mode for minregion=X and minorder=X smoothing (foldfirst/sequence) [foldfirst]
iuscoredir=PATH : Path in which to save protein acc.DISORDER.txt score files for re-use []
discalculate=T/F: Whether to try to calculate disorder if existing score not loaded [True]
md5acc=T/F : Whether to use md5sum hexdigest hashing of sequence in place of accession numbers [True]

System Settings

iupath=PATH : The full path to the IUPred executable [iupred]
anchor=PATH : Full path to ANCHOR executable []
filoop=INT : Number of times to try connecting to FoldIndex server [10]
fisleep=INT : Number of seconds to sleep between attempts [2]
iuchdir=T/F : Whether to change to IUPred directory and run (True) or rely on IUPred_PATH env variable [False]

History Module Version History

    # 0.0 - Initial Compilation.
    # 0.1 - Added parsing of disorder from name as an option instead of disorder prediction
    # 0.2 - Added Folded tuple as well as disordered
    # 0.3 - Added PrintLog opt attribute
    # 0.4 - Added option for correct use of IUPred_PATH environment variable
    # 0.5 - Added Minimum length of an ordered/disordered region
    # 0.6 - Added ANCHOR prediction.
    # 0.7 - Added globProportion calculation.
    # 0.8 - Added makeRegions() method.
    # 1.0.0 - Added random disorder function and elevated to v1.x as fully functional for SLiMSuite
    # 1.1.0 - Added strict option for disorder method selection. Added minorder=X.
    # 1.2.0 - Added saving and loading scores to IUScoreDir/.
    # 1.3.0 - Switched default behaviour to be md5acc=T.
    # 1.4.0 - Fixed up disorder=parse and disorder=foldindex.
    # 1.5.0 - Added iupred2 and anchor2 parsing from URL using accnum. Made default disorder=iushort2.
    # 1.5.1 - Fixed iupred2 URL generation error.
    # 1.5.2 - Changed iupath default.
    # 1.6.0 - Initial Python3 code conversion.