CDS2SFS

Calculates site frequency spectra of synonymous and nonsynonymous sites for coding alignment

Author:

Sarah B. Kingan

University of Rochester

17 March 2014

last update 26 March 2015

Input:

Onne or more files containing multiple alignments of CDS in fasta format.

Formatting requirements for alignment:

1. sequence must be coding (length is multiple of 3, no premature stop codons)
2. sequences may include terminal stop codon or not, but must be uniform across sequences
3. all sequences must be the same length
4. single outgroup sequence identified with 'outroup' in header line
5. files must have ".fa" extension

#Output:

1. STOUT tab delimited text with the following fields (where N is the number of ingroup samples):

    filename_Syn S1 S2 ... SN
    
    filename_NonSyn S1 S2 ... SN
   

2. STDERR progress reports, alignment information, details about excluded sites

Example Calls:

SINGLE FILE:

CDS2SFS.pl mydata.fa > mydata_sfs.out 2> mydata_sfs.err

ALL FILES IN DIRECTORY:

CDSS2SFS.pl > mydata_sfs.out 2> mydata_sfs.err

Exclusion of sites/handling of missing data:

1. codons with any character other than AGCT in any sample are excluded in all samples
2. codons with sites that have more than 2 states in the ingroup sequences are excluded (violation of infinite sites model)
3. codons where ancestral state of the ingroup sequences cannot be inferred are ommited
4. codons where the ingroup has more than two states are excluded

Revision History:

21 April 2014, modified to print out per gene stats - Sarah B. Kingan

22 January 2015, modified to be generalized to any dataset - Anthony J. Geneva

23 January 2015, modified to take either a single file or all files in directory - Sarah B. Kingan

27 January 2015, modified to accept files with no terminal stops and to include codons where neither ingroup codon matches outgroup codon but ancestral codon can be inferred. - Sarah B. Kingan

26 March 2015, commented and reformated for submission to github - Sarah B. Kingan