Coding Potential Calculator (CPC) is a Support Vector Machine-based classifier to assess the protein-coding potential of a transcript (i.e whether a cDNA/RNA transcript could encode a peptide or not) based on six biologically meaningful sequence features. It takes nucleotide FASTA sequences as input, and generate output about the coding status and the "supporting evidence" for the sequence.

1. NCBI BLAST package: a local version could be downloaded from http://www.ncbi.nlm.nih.gov/blast/

2. A relatively comprehensive protein database. UniRef90 and NCBI nr should be both okay.
   The database should be named as "prot_db", and put under the data/ subdir.

1. Unpack the tarball:

   tom@linux$ gzip -dc cpc-0.9-r2.tar.gz | tar xf -

2. Build third-part packages:

   tom@linux$ cd cpc-0.9-r2 tom@linux$ export CPC_HOME="$PWD" tom@linux$ cd libs/libsvm tom@linux$ gzip -dc libsvm-2.81.tar.gz | tar xf - tom@linux$ cd libsvm-2.81 tom@linux$ make clean && make tom@linux$ cd ../.. tom@linux$ gzip -dc estate.tar.gz | tar xf - tom@linux$ cd estate tom@linux$ make clean && make

3. Format BLAST database, named it as "prot_db", and put under the cpc/data/.

   tom@linux$ cd $CPC_HOME/data tom@linux$ formatdb -i (your_fasta_file) -p T -n prot_db

4. Run the predict

   tom@linux$ cd $CPC_HOME tom@linux$ bin/run_predict.sh (input_seq) (result_in_table) (working_dir) (result_evidence)

========= See the website for tutorial and more details. (http://cpc.cbi.pku.edu.cn)

Contact: cpc@mail.cbi.pku.edu.cn