MSS: Whole Genome Alignment using a
Mutation Sensitive Approach

Home   -   Download   -   Experimental Results   -   Related Links

Download

Please download the following single file for the software. (Total size: 284KB)

Linux/Unix version: MSS (mss.tar.gz)

The package contains all the required programs and scripts. Currently, it only has the Linux/Unix version. Installation of the software requires the stardard development tools "make" and "g++". If you want to have graphical outputs, you should have gnuplot installed.

Installation

To install the software, download the file mss.tar.gz and follow the steps below.

  1. Unzip the file by gunzip mss.tar.gz. You will get mss.tar.
  2. Extract the file by tar -xzvf mss.tar. You will get 3 directories (data, sample and source) and two files (Makefile and mss.sh)
  3. Compile the programs by make.

Input File Formats

The input genomes should be stored in files in FASTA format.

Sample Input Sequence Files: Bm.fasta, Xc.fasta

Normal Usage of the Software

For normal users, the default values of the software should be fine for most of the cases (please see the section at the bottom for advanced usage of the software). For the normal usage, the only parameters needed for the program are:
  • The two input file names.
  • The flag for aligning sequences in DNA level or in translated protein level.
Sample run:
  1. To align in DNA level, please add the '-D' flag.
         ./mss.sh -D Bm.fasta Xc.fasta
  2. To align in translated protein level, no flag is needed.
         ./mss.sh Bm.fasta Xc.fasta
Note: Be sure to run mss.sh in the same directory as the FASTA files, otherwise errors may occur.

Output Formats

The software produces outputs in two formats. One is a graphical display of the potential gene regions (in ps format), and the other is in MUM file format.

  • Graphical Display of Potential Gene Regions

    One postscript file will be generated by the default setting of the software.

    Sample Output ps File: BmXc.mum2.ps

    Each red line represents a MUM pair with same orientation in both genomes, while each green line represents a MUM pair with different orientations.



  • MUMs (or clustered MUMs)

    The MUM file is primarily a 6-column list. Each line in the file represents a MUM and each of the six columns correponds to an attribute of that MUM. Lines are in the format of weight s1 s2 e1 e2 sign, where

    • weight is the weight of the MUM. Note that the weight may not equal to the length of the MUM.
    • s1 and e1 are the start and end positions of the MUM in sequence 1.
    • s2 and e2 are the start and end positions of the MUM in sequence 2.
    • sign is the sign of the MUM.
    Note: The output file should be sorted by s1.

    Sample Output MUM File: out.h16.2

Advanced Usage of the Software

Now we will explain the meanings of all the different parameters here, since there are quite a number of parameters needed for the software. We expect only advanced users will use these options, since default settings work fine in most cases.

  • Path of the input files
    Note: Currently our software does not support wildcards or pathname expansion. So typing seq*.txt will not map to seq1.txt or seq2.txt. Instead, a File Not Found error will be reported.


  • Parameters of the programs, including
    • General Options:
      1. -o pfxSet the prefix of the output files (default: "out")
      2. -DAlign genomes in DNA level (default: align in protein level)
      3. -l numSet the minimum length of an exact match for the 1st filtering process (default: 9)
      4. -f numSet the gap distance for the 1st filtering process (default: 6000)
      5. -L numSet the minimum length of an exact match for the 2nd filtering process (default: 12)
      6. -F numSet the gap distance for the 2nd filtering process (default: 6000)
      7. -w numSet the weight-factor for the output of MaxMinCluster (default: 1.4)
      8. -UDo not generate graphical output (default: graphical outputs are genereated)
      9. -hShow all possible options
    • MaxMinCluster Options:
      1. -MUse the gap model as the distance requirement. (default: the fraction model is used)
      2. -k numSet the noise level of the clustering algorithm (default: 3)
      3. -g numSet the maximum gap of matches between a cluster (default: 2000)
      4. -s numSet the minimum size of a cluster (default: 8)
      5. -d numSet the gap difference for the fraction model (default: 5)
      6. -r numSet the gap_ratio for the fraction model (default 0.1)
    • MSS Options:
      1. -t numSet the gap_threshold of matches between a cluster (default: 2.7)
      2. -m numSet the maximum number of mutations allowed (default: 40)