More Errors; add_qiime_labels.py

[Samuel Major]

It seems to be that I am not able to make QIIME give me any kind of quantitative insights into microbial ecology. 

I had some errors last week that was giving me error codes UnicodeDecodeError: 'utf8' codec can't decode byte 0x9c in position 42: invalid start byte. Tony Walters had suggested that this may be due to non ASCI characters, but the code he provided (parse_nonstandard_chars.py) to remove those characters didn't seem to work (or I saved his code in the wrong location, as I don't know where to save codes for python; I had assumed within the qiime scripts).

So now I have all this data that I generated from MacQiime that will not work on Qiime's Virtual Box... as last ditch effort, I decided to try to start from the beginning by using the trimmed and paired Illumina data I previously used to create all my work within MacQiime: I ran 

add_qiime_labels.py -i Spr16_analysis/seqs -m  MacQiime_Analysis/Possibly_corrupt/Fall_15_map_validation/Fall_15_map_corrected.txt -c InputFileName -o Spr16_analysis/concatenated_seqs

(I sequestered the files that hadn't been working to a file called "Possibly_Corrupt" because nothing has been successful with any codes I've run the past month).

This attempt gave me a new error code I haven't seen before. 

Traceback (most recent call last):

  File "/usr/local/bin/add_qiime_labels.py", line 111, in <module>

    main()

  File "/usr/local/bin/add_qiime_labels.py", line 107, in main

    output_dir, count_start)

  File "/usr/local/lib/python2.7/dist-packages/qiime/add_qiime_labels.py", line 44, in add_qiime_labels

    fasta_files = get_fasta_fps(fasta_dir, fasta_name_to_sample_id.keys())

  File "/usr/local/lib/python2.7/dist-packages/qiime/add_qiime_labels.py", line 123, in get_fasta_fps

    raise IOError("Unable to open %s" % curr_fp)

IOError: Unable to open Spr16_analysis/seqs/

Why is this directory unable to be opened?

Is it possible that all of my data has been corrupted? 

Is it possible that there is something wrong with my qiime virtualbox download? 

Will I be able to recover my data or will I have to start from square 1?

[abir...@gmail.com]

Hi, Samuel,

I think it is a leap to conclude that anything has been corrupted at this point.  The most likely reason that the code wouldn't be able to open the directory in question is that it isn't in the expected location; is Spr16_analysis/seqs located directly within the directory you were working in when you ran the command line above?  If not, that would explain the issue.

However, I would suggest that rather than start over on the virtual box, you follow up on the script Tony wrote for you.  What error did it give when you ran it, and have you posted it to the forum?  If not, could you do so now?

Thanks,

Amanda

[Samuel Major]

Thank you for the response Amanda,

All the files are where they should be. The files within the seqs folder are .fa files, but that shouldn't matter. If there was no directory/file where the machine was looking, shouldn't it say "No file/directory found" or something to that effect?

Also I just got a "command not found" error with Tony Walter's code. I'm not 100% I have the command in the right directory in the qiime machine.Thanks. I'll definitely be sure to post to that forum too.

Thank you

Sam