ComputeNEC: Effective number of codons

In fredysiegrist/statanacoseq: Statistical Analyses of Codon Usage in Custom Genome Sequences

Description

ComputeNEC is a method for measuring synonymous codon usage bias in a gene by estimation of the “effective number of codons” Nc

Usage

ComputeNEC(cds)

Arguments

cds	Coding sequence in reading frame

Details

Based on early codon usage mesures proposed by Wright 1990 and Fuglsang 2004. Original implementation from Alexander Roth 2005-2007, second of several indices for codon usage. Nc is the total number of different codons used in a sequence. The values for Nc range from 20 to 61. Where in the first case only one codon is used per amino acid, and in the latter all posible ones are used. Highly expressed genes should use fewer codons due to selection away from equal use. The here computed overall number of effective codons for a gene is a sum of average homozygosities for different redundency classes.

Value

Numerical value for effective number of codons Nc (by the original definition)

Original code in Darwin

Effective number of codons* (Wright 1990, *Fuglsang 2004). AR (April 2007)

ComputeNEC := proc(d:string)
 cod:=CreateArray(1..64);
 aa:=CreateArray(1..20);
 aviod:={op(AToCodon('$'))};
 count:=0;
 for i to length(d) by 3 do
   c := d[i..i+2];
   if not member(c,aviod) then
     ai:=CodonToInt(c);
     ci:=CodonToCInt(c);
     cod[ci]:=cod[ci]+1;
     aa[ai]:=aa[ai]+1;
     count:=count+1;
   fi;
 od;

 Nc:=0;
 for i to 20 do
   Acods := IntToCInt(i);
   k := length(Acods);
   if k<2 then Nc := Nc + 1; next; fi;
   n := sum([seq(cod[Acods[x]], x=1..k)]);
   S := sum([seq((cod[Acods[x]]/n)^2, x=1..k)]);
   F := (n*S-1) / (n-1);
   Nc := Nc + 1/F;
 od;
 Nc;
end: 

Author(s)

Roth, A.; Siegrist, F. and Cannarozzi, G. M. [e-mail]

See Also

seqinr CUB statanacoseq http://gtrnadb.ucsc.edu/GtRNAdb2/ readstats

Examples

ComputeNEC('ATGTGGTACTCCGACTACGGAGGATAA')
ComputeNEC(c2s(mylist(whatout=1)[[1]]))

fredysiegrist/statanacoseq documentation built on May 16, 2019, 2:44 p.m.