Hi VJ, you only can have one <command> tag in a tool, afaik. You can try to write it with cheetah and concatenate several shell commands with && or you can extend your python script to call your java tool at first step. Cheers, Bjoern

Hi,

I am trying to integrate a tool into galaxy. The tool runs in two parts - 1) Computes correlation and provides an ouput 'txt' file (Java) 2) Takes the previously ouput txt file and outputs a 'pdf' file. (Python)

I am having trouble with the second tool. If I provide a sample 'txt' file as input, it's working fine. But if I run it in galaxy, it is not working (for some reason it is not reading the 'txt' file from the previous output). Below is the code.

#!/usr/bin/python

import sys, os import numpy as np import matplotlib.pyplot as plt

input_file = sys.argv[1] output_file = sys.argv[2]

print 'Number of arguments:', len(sys.argv), 'arguments.' print 'Argument List:', str(sys.argv)

txt_in = input_file + '.txt' mydata = np.loadtxt(txt_in) plt.figure(1) plt.plot(mydata[:,0],mydata[:,1],label = "Feature"); plt.plot(mydata[:,0],mydata[:,2], label = "Variance"); plt.xlabel('Distance from feature(bp)'); plt.ylabel('Score'); plt.title(sys.argv[1]); plt.legend(loc="best") pdf_out = output_file + '.pdf' plt.savefig(pdf_out)

data = file(pdf_out, 'rb').read() fp = open(output_file, 'wb') fp.write(data) fp.close() os.remove(txt_in) os.remove(pdf_out)

Below is the xml code.

<tool id="archtex_massdata_extraction" name="Extract mass data"> <description> for the given BAM file </description>

<command> java -jar Extraction.jar $input_bam_file $ref_filename $ref_filetype $output1 </command> <command interpreter="python"> plot.py $output1 $out_file1 </command>

<inputs> <param name="input_bam_file" type="data" format="BAM" label="Input BAM file" help="Choose an input BAM file"/> <param name="ref_filename" type="data" format="gen,txt,gtf,bed" label="Reference/Coordination file" help="Choose a reference file"/> <param name="ref_filetype" type="select" label="Choose the reference file type"> <option value="custom">Custom</option> <option value="refgene">refGene</option> <option value="GFF">GFF</option> <option value="bed">BED</option> </param> </inputs> <outputs> <data name="output1" format="txt" /> <data name="out_file1" format="pdf" /> </outputs> </tool>

I am forcing the input to be 'tx't and the output to be 'pdf' in the python script. When I run the code, it is not showing any errors but it's not showing any output either. I am able to download the first output 'txt' file but there is no download button option in the right pane of galaxy for the second part.

Any help is appreciated!

Thanks, VJ. ___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: http://lists.bx.psu.edu/

To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/

Apologies if my previous suggestion confused you : http://dev.list.galaxyproject.org/error-while-writing-to-pdf-in-python-tp466... You may first generate the pdf : input_file = sys.argv[1] output_file = sys.argv[2] #txt_in = options.exp_data # + '.txt' ## NOT REQUIRED mydata = np.loadtxt(input_file) plt.figure(1) plt.plot(mydata[:,0],mydata[:,1],label = "Feature"); plt.plot(mydata[:,0],mydata[:,2], label = "Variance"); plt.xlabel('Distance from feature(bp)'); plt.ylabel('Score'); plt.title(sys.argv[1]); plt.legend(loc="best") pdf_out = output_file + '.pdf' plt.savefig(pdf_out) ## Write back to the output file ## You can do a shutil.move too [http://docs.python.org/2/library/shutil.html] ## like this : shutil.move(pdf_out,output_file) data = file(output_file, 'rb').read() fp = open(output_file, 'wb') fp.write(data) fp.close() I tried it locally and it worked. But this is under the assumptio that I am passing your python code the input-file from the Galaxy interface itself . I just created a CSV and ran the tool. The CSV was simple enough : 0 1 2 2 3 6 Saket On 11 July 2013 00:19, vijayalakshmi <vs49@buffalo.edu> wrote:

Hi,

I am trying to integrate a tool into galaxy. The tool runs in two parts - 1) Computes correlation and provides an ouput 'txt' file (Java) 2) Takes the previously ouput txt file and outputs a 'pdf' file. (Python)

I am having trouble with the second tool. If I provide a sample 'txt' file as input, it's working fine. But if I run it in galaxy, it is not working (for some reason it is not reading the 'txt' file from the previous output). Below is the code.

#!/usr/bin/python

import sys, os import numpy as np import matplotlib.pyplot as plt

input_file = sys.argv[1] output_file = sys.argv[2]

print 'Number of arguments:', len(sys.argv), 'arguments.' print 'Argument List:', str(sys.argv)

txt_in = input_file + '.txt' mydata = np.loadtxt(txt_in) plt.figure(1) plt.plot(mydata[:,0],mydata[:,1],label = "Feature"); plt.plot(mydata[:,0],mydata[:,2], label = "Variance"); plt.xlabel('Distance from feature(bp)'); plt.ylabel('Score'); plt.title(sys.argv[1]); plt.legend(loc="best") pdf_out = output_file + '.pdf' plt.savefig(pdf_out)

data = file(pdf_out, 'rb').read() fp = open(output_file, 'wb') fp.write(data) fp.close() os.remove(txt_in) os.remove(pdf_out)

Below is the xml code.

<tool id="archtex_massdata_extraction" name="Extract mass data"> <description> for the given BAM file </description>

<command> java -jar Extraction.jar $input_bam_file $ref_filename $ref_filetype $output1 </command> <command interpreter="python"> plot.py $output1 $out_file1 </command>

<inputs> <param name="input_bam_file" type="data" format="BAM" label="Input BAM file" help="Choose an input BAM file"/> <param name="ref_filename" type="data" format="gen,txt,gtf,bed" label="Reference/Coordination file" help="Choose a reference file"/> <param name="ref_filetype" type="select" label="Choose the reference file type"> <option value="custom">Custom</option> <option value="refgene">refGene</option> <option value="GFF">GFF</option> <option value="bed">BED</option> </param> </inputs> <outputs> <data name="output1" format="txt" /> <data name="out_file1" format="pdf" /> </outputs> </tool>

I am forcing the input to be 'tx't and the output to be 'pdf' in the python script. When I run the code, it is not showing any errors but it's not showing any output either. I am able to download the first output 'txt' file but there is no download button option in the right pane of galaxy for the second part.

Any help is appreciated!

Thanks, VJ.

___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: http://lists.bx.psu.edu/

To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/