TooT-SC.v1.svm.11

This tool predicts the substrate class of a given transporter. The class can belong to one of the following eleven categories [1] nonselective inorganic molecule [2] water [3] inorganic cation [4] inorganic anion [5] organic anion [6] organooxogyn [7] amino acid and derivatives [8] other Organonitrogen compound [9] nucleotide [10] organic heterocyclic [11] miscellaneous.

Details on ChEBI mapping is found on Substrate_ChEBI.csv file.

Input: transporter proteins sequences in Fasta format Output: the substrate class with highest probability, and the probabilities of the other classes.

Installation

git clone https://github.com/bioinformatics-group/TooT-SC.git
cd TooT-SC
wget https://tootsuite.encs.concordia.ca/databases/TooT-SC/v1.svm.11_db.tar.gz
wget https://tootsuite.encs.concordia.ca/datasets/TooT-SC/v1.svm.11_datasets.tar.gz
tar -xzf v1.svm.11_db.tar.gz
tar -xzf v1.svm.11_datasets.tar.gz
rm v1*

Use curl if you don't have wget.

FOLDERS

There are a number of folders that support the running of TranCEP and its outputs.

dataset/transporter_substrate_class

Contains both the training and the independent testing dataset if fasta format, the identifiers of independent testset is found under substrate_classes_indep.csv. The models were trained using the training dataset =(all dataset - independent testset)

Working Directory

When run, the tool will create a work directory, common among all the TooT Suite. TooT-SC will create a directory named TooT-SC which will hold intermediate files created during the running of these tools. The top-level working directory contains the homology details needed to extract the features. Details of the Blast hits for each sequence are found here, one directory per sequence.

The TooT-SC directory will also contain a Compostions folder which contains the extracted MSA_TPAAC features of the test set

The working directory can be quite large, so you may want to use -work to specify a better location for it if you're short on space, and be sure to clean it out from time to time.

db

The database to be used when performing BLAST.

By default, if you tar -xzf this adjacent to the source folder, it should create a db folder in the default location that this tool checks. You may also specify the location manually via -db. This should be the Swissprot Oct. 2018 database used in ther related publication.

src

The scripts needed to use the tool.

HOW TO USE

• This tool requires that BLAST be pre-installed
• Usage: src/TooT_SC.R -query=<input> [-out=<outdir>] [-db=<path to db>] [-work=<Workdir>] [-TooTSC=<TooTSCdir>]
• <input> is your sequence input file in fasta format
• <out> is the output directory where you want the predicted results, formatted as csv
• <db> is the directory where the database is stored
• <Workdir> is the directory where intermediate files will be stored after each run
• <TooTSCdir> is the directory where the base TooT-SC files are located
• MSA_PAAC features of each sequence in the test set are found under the working directory Compositions/MSA_PAAC.csv