In lukembrowne/sgsR: Spatial genetic structure analyses in R

sgsR

sgsR is a package for calculating spatial genetic structure in R. The aim is to implement analyses, similar to those found in SPAGeDi and GenAlEx, that estimate the degree of spatial autocorrelation in genetic data.

Some key features of sgsR are:

1. Calculating relatedness among individuals based on set distance intervals
2. Conducting permutation tests
3. Creating spatial autocorrelation plots
4. Reading and converting from SPAGeDi and genepop data format

---

sgsR is very much still in development, and certainly contains bugs, though I've done my best comparing results from sgsR to results from SPAGeDi. If you're interested in contributing, have any comments or suggestions for features you'd like to see, please get in touch via email - [e-mail]

Here's an example of a typical workflow...

• Input data into the sgs data structure using createSgsObj()
• Set desired distance intervals and number of permutations
• Run SGS analysis with sgs(), currently configured to run the kinship coefficient of Loiselle et al. 1995
• use plotSgs() to produce an autorcorrelation plot

library(sgsR)

## Simulate genetic data
    Nind = 100
    Nloci = 10
    Nallele = 10
    n =  Nind * 2 # Number of gene copies

    ## Initialize data frame
    dat <- data.frame(id = 0:(Nind-1))
    dat$x = runif(Nind, 0, 100)
    dat$y = runif(Nind, 0, 100)

    ## Simulate Random genetic data
    for(loci in 1:Nloci){
      loci_name_a = paste("Loc", loci, "_A", sep = "")
      loci_name_b = paste("Loc", loci, "_B", sep = "")
      dat[loci_name_a] <- sample.int(Nallele, Nind, replace = TRUE)
      dat[loci_name_b] <- sample.int(Nallele, Nind, replace = TRUE)
    }

## Convert to sgsObj
sgsObj = createSgsObj(sample_ids = dat$id, 
                      genotype_data = dat[, 4:(Nloci*2 + 3)],
                      ploidy = 2,
                      x_coords = dat$x, 
                      y_coords = dat$y)

# Display genetic data
head(sgsObj$gen_data)


## Run analysis
distance_intervals = seq(10, 110, 10) # Set distance intervals

out1 = sgs(sgsObj = sgsObj, distance_intervals = distance_intervals, nperm = 99)




## Plotting results

## Solid line is Fij estimate for each distance class
## Dashed lines are the 2.5 % and 97.5 % quantiles of the permuted values
plot(out1)

# Summary of information on distance classes
out1$di

# Summary of information on estimated Kinship coefficient for each distance class (columns)
round(out1$fij_obs, 3)

lukembrowne/sgsR documentation built on May 21, 2019, 8:58 a.m.