NORSnet is a neural network based method that focuses on the identification of unstructured loops. NORSnet was trained to distinguish between very long contiguous segments with non-regular secondary structure (NORS regions) and well-folded proteins. NORSnet was trained on predicted information rather than on experimental data. Therefore, it was optimized on a large data, which is not biased by today's experimental means of capturing disorder. Thus, NORSnet reached into regions in sequence space that are not covered by the specialized disorder predictors. One disadvantage of this approach is that it is not optimal for the identification of the "average" disordered region.
Installation with aptitude (Debian, Ubuntu, etc.)
- If you have not done so until now, add the rostlab repository to the list of your syanptic package manager. This is how it's done: Debian_repository#sources.list.d
- aptitude update
- aptitude (search for rostlab keyring and install by marking the package with a '+' and hit 'g' twice to install)
- aptitude update (to determine all rostlab packages to install)
- aptitude install norsnet. Here's a step by step guide Debian_repository#Installing_a_package_step_by_step
In order to properly run NORSnet you will need to install the following packages. These will be used to create the required NORSnet input files.
- BLAST - The popular sequence alignment program.
- PROFphd - Secondary structure and solvent accessibility predictor. Package also provides useful conversion utilities see Generating an HSSP Input File .
- PROFBval - predicts backbone B-values from amino-acid sequence.
Generating Input Files for NORSnet
NORSnet takes 3 input files:
- sequence_file - file containing protein amino-acid sequence in fasta format.
- filtered_hssp_file - PSI-BLAST alignment profile file.
- rdbProf_file - Secondary structure and solvent accessibility prediction by PROFphd in rdb format.
- profbval_file - predicted B-values by PROFBval.
Generating an HSSP Input File
Here is a protocol for converting an alignment to an HSSP profile.
Generating an rdbProf File
Generating an rdbProf file requires generating an HSSP input file. The HSSP input file is used as input to the prof program.
prof <filtered_hssp_file> both numresMin=17 nresPerLineAli=60 riSubSec=4 riSubAcc=3 fileRdb=<rdbProf_file>
Generating a PROFBval File
NORSnet takes a raw neural network output. In order to produce this it needs to be set to mode 5. The mode is set by the 7th command line parameter. For more information see the PROFBval page.
profbval <sequence_file> <rdbProf_file> <filtered_hssp_file> [protein name] <profbval_output_file> 9 5
Please see the NORSnet man page: