Skip to content

JiscaH/sequoia

Repository files navigation

sequoia

For the website, go to: https://jiscah.github.io/

General

Sequoia reconstructs multi-generational pedigrees from SNP data, as described in Pedigree reconstruction using SNP data: parentage assignment, sibship clustering, and beyond ( http://onlinelibrary.wiley.com/doi/10.1111/1755-0998.12665/full ). It performs parentage assignment (parent genotyped) and sibship clustering (parent not genotyped), as well as grandparent assignment to link sibship clusters to the rest of the pedigree.

The R package also includes a range of other functions, amongst others to check if a pedigree is accordant with the SNP data, and to compare two pedigrees.

For successfull pedigree reconstruction, sequoia ideally requires data on a few hundred SNPs with high MAF, low missingness, low genotyping error rate, and in low LD, but real-world data usually works well too. In addition it is useful to have sex and birth year information of as many individuals as possible.

Download & installation

The latest stable and thoroughly bug and performance checked version of the package is available from CRAN, and can be downloaded installed using (in R)

install.packages("sequoia")

The latest not quite so thoroughly tested version can be downloaded and installed using

remotes::install_github("JiscaH/sequoia")

Note that this package requires compilation, as the bulk of the algorithm is written in Fortran. A pre-compiled .zip binary file of the development version (for Windows and the current R version only) can be found in sequoia_archives. You can download this to your hard drive, and then install using

install.packages("C:/file/to/path/sequoia_2.3.3.zip",  repos = NULL)`

Sometimes it seems necessary to first rename the file to 'sequoia.zip', if R can't find the package back after installation, and sometimes you need to turn R off & on before it works.

The archives also has the source files and pre-compiled Windows and MacOS binaries of old versions that used to be on CRAN, as well as source files of old versions that never made it to CRAN.

Running

The function to perform pedigree reconstruction is also called sequoia():

# load the package
library(sequoia)  

# load example data 
data(SimGeno_example, LH_HSg5, package="sequoia")  

# run pedigree reconstruction
SeqOUT <- sequoia(GenoM = SimGeno_example, 
                  LifeHistData = LH_HSg5, 
                  Err = 0.005,   # genotyping error rate
                  Module="ped", 
                  quiet="verbose", 
                  Plot=TRUE)
# the result is a list with the pedigree, run parameters, 
# and various other elements.                 

# graphical summary of results
SummarySeq(SeqOUT)

Detailed instructions on how to use the package are available in the vignette (vignettes/vignette-main.pdf), available in R via vignette("sequoia") or help(package="sequoia").

Not R

A stand-alone Fortran version can be found in sequoia_notR . It does not require the genetic data to pass through R, which may be nearly impossible for very large datasets. Note that it does not offer all the features of the R package, as detailed in its manual. Its version will typically be functionally identical to the latest R package development version.

More R functions

A collection of additional sequoia-related R functions can be found in sequoiaExtra.