Coding Potential Calculator (CPC) error

Question

[Janey]

Hi

In the first phase, I ran this command for Coding Potential Calculator (CPC):

formatdb -i X.fasta -p T -n prot_db

Then, I got 3 files " prot_db.psq, prot_db.pin, prot_db.phr". I put these 3 files and X.fasta in data/ sub-directory.

In the second phase, when I run bellow command in bin/ sub-directory:

run_predict.sh Query.fasta

I get this error:

Permission denied

Help me please

[grant.hovhannisyan]

Maybe try first chmod 777 run_predict.sh (this opens all read/write permissions), and then run the script like ./run_predict.sh Query.fasta result_in_table outpur_dir result_evidence

[Janey]

Hi grant

After using chmod 777 run_predict.sh and then running this command:

./run_predict.sh Query.fasta result_in_table outpur_dir result_evidence

I get this error:

./bin/run_predict.sh: line 121: outpur_dir/blastx.feat1: No such file or directory
./bin/run_predict.sh: line 123: outpur_dir/ff.feat: No such file or directory
./bin/run_predict.sh: line 128: outpur_dir/blastx.feat: No such file or directory
./bin/run_predict.sh: line 130: outpur_dir/ff.fa: No such file or directory
cat: outpur_dir/ff.fa1: No such file or directory
./bin/run_predict.sh: line 140: outpur_dir/blastx.lsv: No such file or directory
./bin/run_predict.sh: line 144: outpur_dir/ff.lsv: No such file or directory
./bin/run_predict.sh: line 147: outpur_dir/test.lsv: No such file or directory
./bin/run_predict.sh: line 148: outpur_dir/test.lsv.scaled: No such file or directory
./bin/run_predict.sh: line 155: outpur_dir/test.svm0.stdout: No such file or directory
cat: outpur_dir/test.svm0.predict: No such file or directory
./bin/run_predict.sh: line 169: outpur_dir/blastx.index: No such file or directory
cat: outpur_dir/blastx.feat: No such file or directory
Can't open BLAST TABLE file (outpur_dir/blastx.table): No such file or directory at ./bin/generate_plot_features.pl line 18.

[grant.hovhannisyan]

Might sound silly, but 1. Do you run all commands in the same directory? Try to do so 2. Try to replace outpur_dir by . (Current directory)

[Janey]

I did all the work you said but still got the same error

[h.mon]

It is about time you try CPC2, then. According to its authors, it runs ∼1000 times faster than CPC1 and exhibits superior accuracy compared with CPC1. Also, CPC1 depends on legacy-blast, which has been abandoned by NCBI for years, in favour of blast+.

[grant.hovhannisyan]

Seems strange. You have blast+, right?

[h.mon]

I think CPC uses legacy-blast, not blast+.

[grant.hovhannisyan]

I personally run it with blast+ installed.

[Janey]

yes, I have blast+ (ncbi-blast-2.7.1+). What do you think of using Blast directly instead of CPC?

[grant.hovhannisyan]

Here is a chunk from manual: c. 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)

So basically try to do formatdb in cpc/data directory, but run_predict.sh in cpc directory. Regarding your question about using only blast, so you will need to calculate coding potential from blast hits, which I think is complicated and basically is inventing the wheel:) Hope this helps

[h.mon]

You know formatdb is a command from legacy-blast, right? Also, run_predict.sh has the following line:

APP_BLAST=`which blastall 2> /dev/null`

This is the legacy-blast executable, and there is no alternative command to use blast+ inside the script.

[Janey]

I got a similar error again.

[Janey]

Hi grant

Finally my problem was solved when made outpur_dir directory in cpc-0.9-r2/bin directory. I got 3 file result_in_table, result_evidence.orf and result_evidence.homo.

thank you

[grant.hovhannisyan]

I'm glad to hear that.

[Ayeh]

Dear Janey

I have similar error like you but I couldn't solve this problem. I would appreciate to help me for solving it

best regards

[h.mon]

Maybe you should consider using CPC2.

[Janey]

CPC2 release temporary package that, in my opinion, is not reliable.