• Resources
    • Request Access
    • Manage Existing Project
    • Compute Resources
    • Software Packages
    • Partner Program
  • Documentation
    • Get Started
    • Log In
    • Transfer Files
    • Storage
    • Compiling
    • Running Jobs
  • Support
    • Contact Us
    • Training and Events
    • Links for Learning
  • Cluster Status
  1. OIT HPC
  2. Software
  3. FALCON

    Falcon



    • A sample Falcon job
    • A sample Falcon Unzip job

      Falcon is a heirarchical genome assembly process. It uses the following steps for generating a genome assembly from a set of sequencing reads:

      1. Raw sub-reads overalapping for error corrections.
      2. Pre-assembly and error correction.
      3. Overlapping detection of the error corrected reads
      4. Overlap filtering
      5. Constructing a graph from overlaps
      6. Constructing contig from graph
      You can copy a working example downloaded from falcon github site.

    • A Sample Falcon job This sample run is too large for /home, but you should be able to run it in your group /share directory.

      cp /usr/local/apps/samples/falcon.tar . 
      tar xvf falcon.tar 
      cd falcon 
      

      Then typing

      bsub < bfalcon 
      

      should run the example which relies on the fc_run_ecoli_local.cfg which in this case used 64 GBytes of RAM and 16 cores. The bfalcon file consists of

       
      #!/bin/ksh
      #BSUB -R span[ptile=16]
      #BSUB -n 16
      #BSUB -q single_chassis 
      #BSUB -W 2000
      source /usr/local/apps/falcon/falcon.ksh
      fc_run.py fc_run_ecoli_local.cfg
      #BSUB -o out.%J
      #BSUB -e err.%J
      

      The only files needed were the bfalcon and run_ecoli_local.cfg (along with input.fofn and the fasta files it points to), the other files and subdirectories were created running falcon. To run your own problem you'll need to modify the .cfg file.

      A main problem in running the job was successfully specifiying the input files. The "file not found error" was not in the err.xxxxx file or in all.log, but in 0-rawreads/pwatcher.dir/stderr. I had to copy input.fofn (specified in the .cfg file) to the directory in which falcon was searching. Another problem was that the .sh files in 0-rawread were not created as executable.

      cd 0-rawreads
      chmod +x *.sh 
      

      made the .sh files executable. This version of falcon links to python-2.7.13.

      For more documentation on this example, see e coli falcon tutorial.

    • A Sample Falcon Unzip job

      Falcon unzip is also part of the falcon tool chain, using files produced by a run of falcon. In the same directory as above (or tar up the directory /home/gwhowell/apps/samples/falcon/greg200k-sv2 and "tar xvf" to unpack it).

       
      cd greg200k-sv2
      rm -rf  0-rawreads
      rm -rf 1-preads_ovl
      rm -rf 2-asm-falcon
      rm -rf 3-unzip
      

      Then

       
      bsub < bfalcon
      

      will run falcon and recreate the first 3 of the above directories (log is logfalcon1 ). Then

       
      bsub < bunfalcon
      
      gives a run of falcon unzip, with log in all.log

      This version used the default fc_unzip.cfg.

      Last modified: April 02 2024 13:40:32.

      Copyright © 2025 · Office of Information Technology · NC State University · Raleigh, NC 27695 · Accessibility · Privacy · University Policies