LinkNe

Overview

LinkNe is a program for estimating effective population size from LD at loci with known linkage relationships. A detailed description of the program can be found in:

Hollenbeck CM, Portnoy DS, Gold JR (2016) A method for detecting recent changes in contemporary effective population size from linkage disequilibrium at linked and unlinked loci. Heredity, 117, 207–216. [link]

Input

Requirements:

1. Genepop file of diploid genotypes

   • the file should contain only one population
   • only the three-character allele format is currently supported
   • missing genotypes are indicated with '000000'

2. Recombination rate estimates for at least a subset of the genetic markers. Two formats are allowed:

• A genetic map with the following format (note that map positions are in cM and the header is required):

    locus chromosome  position
    loc1  1 5.0
    loc2  1 15.0
    loc3  2 50.5

• A square matrix of recombination rates (note that recombination rates are in Morgans). Example (equivalent to the map above):

    loc1  0 0.05 0.50
    loc2  0.05 0 0.50
    loc3  0.50 0.50 0

Output

Ne.out

The columns in the Ne.out file correspond to the following:

• MIDPOINT_C: The midpoint recombination rate of the bin
• MEAN_C: The mean recombination rate of the bin
• NE: The point estimate of effective population size
• PARA_95_LOW: Lower bound of 95% parametric confidence interval
• PARA_95_HIGH: Upper bound of 95% parametric confidence interval
• PAIRWISE: Number of pairwise locus comparisons used in the estimate
• S: Mean sample size
• CV: Coefficient of variation - from Hill (1981)
• EMP_95_LOW: Lower bound of 95% confidence interval based on CV from Hill (1981)
• EMP_95_HIGH: Upper bound of 95% confidence interval based on CV from Hill (1981)
• MEAN_R_SQ: Mean total r-square
• MEAN_EXP_R_SQ: Mean expected r-square
• R_SQ_DRIFT: R-square attributable to drift
• R_SQ_DRIFT_LOW: Lower bound of 95% confidence interval for r-square drift
• R_SQ_DRIFT_HIGH: Upper bound of 95% confidence interval for r-square drift

Usage

perl LinkNe.pl -i input_file -map linkage_map [options]


 Options:

     -i, --infile
             Genepop input file - currently only supports three-character alleles where '000000' indicates a missing genotype

     -map    Linkage map in tab-separated format (with header):

            <locus> <chromosome>  <position>

            Example:

            locus chromosome  position
            loc1  1 5.0
            loc2  1 15.0
            loc3  2 50.5

            Position values should be in centiMorgans

     -m, --matfile
             Matrix of recombination frequencies for loci

     -o, --outfile
             Outfile

     -b, --binsize
             Size of bins (in Morgans) for estimating Ne [Default: 0.05]

     -t, --timebin
             Bin pairwise estimates of LD by generations (measured by 1 / 2c)
             rather than by recombination frequency (c) itself

     -a, --allele_cutoff
             Cutoff frequency for excluding rare alleles from the analysis
             [Default: 0.05]

     -e, --rec_cutoff
             exclude locus pairs below a specified recombination faction
             [Default: No cutoff]

     -c, --no_bias_corr
             Turn off correction of expected r2 values using the bias
             correction of Waples (2006) - not recommended except for
             experimental use

     -v, --moving_avg
             Compute a moving average for effective size relative to
             recombination rate. Must specify a window size and interval for
             moving the window.

     -w, --window
             Size of sliding window for calculating moving average (in
             Morgans) [Default: 0.01]

     -n, --interval
             Distance the sliding window moves for each calculation (in
             Morgans) [Default: 0.005]

     -s, --save
             Saves relevant pairwise data to a file for later recalculation
             of the moving average

Dependencies

The following perl modules are required for running LinkNe:

Data::Dumper
Getopt::Long
Pod::Usage
Statistics::Distributions