Francisco C. Ceballos 2023-11-15
- Package Overview.
- Processing .hom files
- Population Genetics.
- ROH islands and Regions of Heterozygosity.
The analysis of autozygosity through the Runs of Homozygosity (ROH)
–contiguous regions of the genome where an individual is homozygous
across all sites– is an alternative and relatively new approach to
answer many important questions in modern genetics, including population
history and the genetic architecture of complex traits.
This
package provides with several useful functions to represent and analyze
PLINK’s ROH outcomes. The following script in bash is recommended to
search of ROH in humans. The following PLINK’s conditions are optimal
for ROH searching in humans when a minimun of 1.5M SNP is used.
plink \
--bfile path/to/bfile \
--homozyg-window-snp 30 \
--homozyg-window-het 1 \
--homozyg-snp 30 \
--homozyg-kb 300 \
--homozyg-density 30 \
--out path/to/folder
This bash script will generate a .hom file with all the information needed. This package takes that tabular file and process it to obtain different summary outcomes, figures and other features.
This package includes different datasets based on the Human Genome
Diversity
Pproject.
HGDP_hom: .hom file with the PLINK’s ROH analysis of 54
populations belonging to the
HGDP_pops: classification in
populations for all the individuals belonging to the HGDP.
HGDP_cont: classification in continents for all the individuals
belonging to the HGDP
HGDP_het: average FIS estimates for
each individual in the HGD dataset:
HGDP_cl_Africa: PLINKs
.hom file obtained using Human Genome Diversity Panel. Cluster Africa.
HGDP_cl_EastAsia: PLINKs .hom file obtained using Human
Genome Diversity Panel. Cluster East Asia.
HGDP_cl_Europe:
PLINKs .hom file obtained using Human Genome Diversity Panel. Cluster
Europe.
HGDP_cl_MiddleEast: PLINKs .hom file obtained using
Human Genome Diversity Panel. Middle East.
ROHi_HGDP: ROH
islands of all the individuals of the HGDP organized by continental
region.
RHZ_HGDP: Runs of heterozygosity of all the
individuals of the HGDP organized by continental region.
First four functions avaliable in this package are useful to first
process the PLINK’s .hom file.
roh_summ_id: This function
extracts various valuable variables from a PLINK home file. One of the
most important is the genomic inbreeding coefficient or FROH. This is
the proportion of the autosomal genome that is in ROH larger than 1.5Mb.
roh_summ_pop: This function summarize all the outcome of
roh_summ_id() by population.
roh_summ_cont: This function
summarize all the outcome of roh_summ_id() by continent.
roh_summ_factor: This function summarize all the outcome of
roh_summ_id() by any factor.
The size of the ROH correlates with the age of that ROH. It is important
to check the ROH size distribution in oprder to assess the age of the
different ROH present in an individual or in a population.
ROH_class_fig function does precisely this. It creates a figure of
the population’s average Sum of ROH for different length ROH classes.
(0.3MB≤ROH<0.5, 0.5≤ROH<1, 1≤ROH<2, 2≤ROH4, 4≤ROH<8, ROH≥8MB). Raw
data from the figure can be obtained using ROH_class_data
function.
Two distinct and independent biological scenarios can increase homozygosity in natural populations: cultural consanguinity and genetic drift in isolated populations. These two different sources were defined in classical population genetics as systematic inbreeding (denoted by FIS) and panmictic nbreeding (denoted by FST), respectively. Total inbreeding, denoted by FIT, is defined by (1-FIT) = (1-FIS) (1-FST). Panmictic inbreeding occurs in isolated populations, when individuals randomly mate within their own group, with no immigration. Population isolation can be cultural, a consequence of geographical barriers or because of sedentary behavior. Importantly, isolation by itself does not create genomic autozygosity, except when the effective population size (Ne) is small and genetic drift has the strength to remove genetic variability. On the other hand, systematic inbreeding, or cultural consanguinity, has the effect of reducing heterozygosity relative to the expectation under Hardy-Weinberg equilibrium independent of Ne, and thus increasing FIS. High consanguinity (and consequent high FIS), and genetic drift by isolation coupled with low Ne (and consequent high FST) are two independent and non-mutually-exclusive phenomena that can increase overall autozygosity (FIT) in a population. Here we also note that the term endogamy is generally used to describe population isolation, although the term is sometimes used to refer to consanguinity.
- Number of ROH vs Sum of ROH We may assess the origins of the autozygosity using comparisons of NROH>1.5Mb and SROH>1.5Mb. Namely, if an individual displays excess of SROH>1.5Mb relative to NROH>1.5Mb in comparison to non-consanguineous individuals, this suggests autozygosity by consanguinity. If both SROH>1.5Mb and NROH>1.5Mb are high, this suggests drift-driven autozygosity. This approach does not provide a quantitative estimation of the elative contributions of drift versus consanguinity, but only a qualitative assessment. Nevertheless, it is a powerful approach: its inferences on autozygosity patterns in modern human populations’ genomes are consistent with known ethnographic data about consanguineous traditions, and about population size and isolation.Simulations of the number and sum of ROHs, for ROH > 1.5 Mb, calculated for the offspring of different consanguineous matings can be shown, along with the ancient and modern samples. Points with different colors designate offspring of different consanguineous mating: second cousin (green), first cousin (yellow), avuncular (uncle-niece, aunt-nephew, double first cousin) (orange), incest (brother-sister, parent-offspring) (red). Five thousand simulations are represented for each consanguineous mating. Note that this simulation does not include drift, but the degree of right shift can be projected to cases where there exists a non-zero level of autozygosity due to drift.
- FIS vs FROH Population analysis and components of the inbreeding coefficient. Systematic inbreeding coefficient (FIS) versus the inbreeding coefficient obtained from ROH (FROH). In this context FIS is the average SNP homozygosity within an individual relative to the expected homozygosity of alleles randomly drawn from the population and it was obtained using the —het function in PLINK. Diagonal broken line represents FIS = FROH. Horizontal broken line represents FIS = 0. Three different regions can be considered in this plot, delineated by the diagonal, where FIS = FROH, and the horizontal line FIS = 0. 1 Populations close to the diagonal line have a strong component of systematic inbreeding or FIS, which means that the total inbreeding coefficient, FIT, of this population is mainly produced by a deviation from panmixia, in other words, consanguinity. 2 Panmictic inbreeding, caused by genetic drift will be more relevant in populations positioned close to the line FIS = 0. 3 Low Ne, isolation and genetic drift become very relevant when populations have negative FIS. Under this scenario of avoidance of consanguinity and excess of heterozygotes (compared to that expected under Hardy–Weinberg (H–W) proportions), the total inbreeding coefficient of these populations will be provoked by genetic isolation and genetic drift: strong FST. Finding populations in the last region to be considered (where FIS > FROH) does not make much sense under an inbreeding context and according to Wright F statistic. If a population presents with a larger FIS than FROH, other phenomena must be taken into account. Besides inbreeding, natural selection pressure and the Wahlund effect can increase FIS; however, natural selection is an evolutionary force that can change FIS locally in specific genome regions, but not at a whole genome level. The only explanation is the Wahlund effect: a deficiency of heterozygotes and excess of homozygotes generated when subpopulations with different allele frequencies are lumped together.
ROH islands are defined as regions in the genome where the proportion of individuals of a population deviates from the expected under a binomial distribution. These regions have been found to be enriched with protein coding genes under selection. To search for ROHi, a sliding window of 100 kb was used. In every 100 kb genomic window, the number of subjects with ROH was obtained and a binomial test was applied (threshold for significance established at p<2x10-6, corresponding to an adjustment for 25,000 windows). Summary information of these islands can be obtained using the function: rohi_summ_pop_file or rohi_summ_pop_path. Also it is possible to represent these islands: rohi_genome.
RHZ are regions in the genome where no individual in the population has a ROH. In order to only identify informative heterozygous haplotypes, regions that have anomalous, unstructured, high signal/read counts in next generation sequence experiments were removed. These 226 regions, called ultra-high signal artefact regions, include high mapability islands, low mapability islands, satellite repeats, centromere regions, snRNA and telomeric regions (Consortium EP 2012). Summary information of these regions can be obtained using the function: rhz_summ_pop_file or rhz_summ_pop_path. Also it is possible to represent these regions: rohi_rohz_genome.
ROH islands and regions of heterozygosiuty (RHZ) can be used to decipher natural population history as also identify protein coding genes under two different kinds of selection: directional dominance and balancing selection. Indeed, searching for the common and unique ROH islands and RHZ between different ethnic groups can be used as a genetic distance measure, or between an affected and unaffected groups can be used to select associated regions. For that reason, a funcion to find this unique and common ROHi or RHZ is provide by this package: comm_uni. As an example a representation of the proportion of pairwise unique ROH islands are provided for the populations present in the African Genome Variation Project. This figure was published by Ceballos et al. 2018. Since the proportion of unique ROHi can be understood as a genetic distance, a philogenetic dendogram was build.
The figure shows a rooted
dendrogram of the unique ROH islands. The rooted dendrogram was obtained
using optimal leaf ordering or OLO. Af.KS African Khoe and San
populations, Af.S population from southern Africa, Af.E population
from eastern Africa, Af.W population from western Africa, Af.GG
population from the Gulf of Guinea, Mix.AA African-American admix
populations, Mix.HA Hispanic-American admix populations, Europe
European populations, Asia.E populations from eastern Asia
Next step in the ROH analysis is to harvest for the protein coding genes present in the ROHi or RHZ. For that purpose biomaRt package is going to be used. To be able to install this package the following script must be ran:
if (!require("BiocManager", quietly = TRUE))
install.packages("BiocManager")## Warning: package 'BiocManager' was built under R version 4.3.2
BiocManager::install("biomaRt")## Bioconductor version 3.18 (BiocManager 1.30.22), R 4.3.1 (2023-06-16 ucrt)
## Warning: package(s) not installed when version(s) same as or greater than current; use
## `force = TRUE` to re-install: 'biomaRt'
## Installation paths not writeable, unable to update packages
## path: C:/Program Files/R/R-4.3.1/library
## packages:
## foreign, KernSmooth, lattice, mgcv, nlme, rpart, spatial, survival
## Old packages: 'curl', 'dplyr', 'fansi', 'htmltools', 'httr2', 'lme4',
## 'lubridate', 'Matrix', 'plyr', 'pROC', 'rlang', 'stringi', 'stringr', 'utf8',
## 'vctrs', 'VGAM', 'xfun'
Once package biomaRt has been installed it is possible to run the function get_Prot