LaNeta

Statistical method for estimating parameters (e. g. the timing) of two pulse model based on three locus Linkage Disequilibrium and Local Ancestry.

Installation

For performance reasons, we use cython to speed up calculations. For this, you need a working toolchain for building C code (gcc and clang are known to work). Since you are going to build Python extensions, you will need python development headers (e.g. on ubuntu linux the package name is python-dev).

The simplest quick-start cross-platform way is to use conda. To do this, create a fresh conda environment:

$ conda create -n conda_laneta
$ conda activate conda_laneta
$ conda install python=3.9       # or other python version of your choice (any `python >= 3.7` should work).

Note that we install python with conda, inside the conda environment. Mixing system-install python and conda may lead to build- or runtime errors.

Get the source code --- here we clone it from GitHub

$ git clone https://github.com/Genomics-HSE/LaNeta.git
$ cd LaNeta

Then build the package:

$ python -m pip install .

That's it!

Quick start

We use .vcf files for admixed and two source populations and simple txt .map and .pop files with morgan units:

To specify all necessary use these parameters: -vcf vcf/dir.vcf.gz -m morgans.map -p populations.pop

Also you need to specify which populations from .pop are admixed and source:

-p0 ADM - admixed population

-p1 SRC1 - admixed 1 time

-p2 SRC2 - admixed 2 times

---

.map format (make sure that there are no duplicates in this file!):

  CHR_NAME VAR_ID POS_PB POS_MORGANS

.pop format:

  SAMPLE_ID POPULATION

---

Example:

python laneta.py -b 0.01 -vcf mer.vcf.gz -p populations.pop -m morgans.map -p0 CLM -p1 YRI -mt 0.94 -jk -nmt

If you specify only one source population, admixed population is separated into two equal-sized groups. These groups are used as the admixed and the missing source population.

All settings

bracket parameters

-b sets the distance between centres of brackets for FFT (default is 0.01).

-r sets the radius of brackets for FFT (default is a half of b).

files

-vcf specifies .vcf file that contains data for all populations.

-p .pop file that contains samples with indicated population.

-m .map file with chromosome name(1-22), var id, var position(bp), var position(cm).

-p0 name of the admixed population in population .txt file.

-p1 name of the first source population in population .txt file.

-p2 name of the second source in popultion .txt file.

two pulse parameters

-mt - total admixture proportion of the second source population. Generally used for estimating allele frequencies for missing source population, however by default it is also used for estimating parameters, to avoid this use -nmt flag.

You can set parameters fixed by providing a values with next keys:

-m1, -m2 used for setting adm. proportions. of the second source population.

-t1, -t2 used for setting adm. times.

cm parameters

-min and -max specifies min and max genetic distance for estimations (in cantimorgans).

flags

-ht if you use haplotype data instead of genotype.

-jk for calculation of confidence intervals using jackknife by leaving out each chromosome.

-nmt if you want mt to be used only for estimating frequencies from missing source population and not for estimating parameters.

Output

Output is a tab delimited list of parameters: t1 - time between two admixture events, t2 - time to the most recent admixture event, m1 - admixture proportion of the second source population for the first admixture event, m2 - admixture proportion of the second source population for the most recent admixture event.

If jackknife used, output have additional lines:

jk bias

jk var

jk 95% conf. interval

Data preparation for analysis

Files which are required for data preparation are places in utilites folder. You need your vcf.gz file and its index file, also check if bcftools, vcftools and plink are installed.

1. Execute get_interpolation_files.sh to get recombination maps for future interpolation. It will be in folder map.
2. To execute preparation.sh you will have to enter the name of your file without .vcf. As a result, you will have .txt file with 4 columns: CHROM ID POS GEN_POS. The forth column refers to morgan units.
3. Repeat with all populations you have
4. Merge all your population file by running

  bcftools merge pop1.vcf pop2.vcf -Oz -o merged.vcf.gz
  plink --vcf merged.vcf.gz --chr 1-22 --snps-only --recode vcf --out filtered_merged
  bcftools view -m2 -M2 -c20 -v snps -Oz -o filtered_merged.vcf.gz filtered_merged

filtered_merged.vcf.gz is the file for your further analysis.