Generating the data for analysis
The first step in exploring genomic prediction is to generate data where we know the answer so that we can compare our estimate from the data with the truth. generate_data()
does this by following these steps.
Generate an genotype at each locus for an individual. By default, generate_data()
generates genotypes at 20 loci, but you can change that when you call it by changing n_loci_total
in the call. The genotype at each locus is generated randomly assuming that the genotypes are in proportions 0.25, 0.5, and 0.25 for one homozygote, the heterozygote, and the other homozygote respectively.
Generate a phenotype for this individual.
- Calculate the genotypic mean at each locus from the specified allelic effect and the genotype at that locus. By default the allelic effects are 1.0, -1.0, 0.5, -0.5, and 0.25 at loci 1-5 and 0.0 at the remaining 15 loci. You can change that by changing
effect
when you call generate_data()
.
- Generate a phenotypic contribution each locus as a random variable having a normal distribution with the genotypic mean and a specified standard deviation. By default the standard deviation is 0.2. You can change that by changing
s_dev
when you call generate_data()
.
- Sum the phenotypic contribution across all of the loci.
Repeat #1 and #2 until you have the number of individuals in the sample data set that you want. By default that’s 100, but you can change that by changing n_indiv
when you call generate_data
.
generate_data()
returns the results of the simulation as a data frame with each individual on a row and with the phenotype in the first column and the genotypes at each locus in the remaining columns.
NOTE: I’m using set.seed()
to ensure that you get the same sequence of random numbers that I do if you run this code on your own. You can delete that line or comment it out if you want to see what happens with different randomly generated data sets.
set.seed(12345)
generate_genotypes <- function(n) {
x <- rmultinom(n, 1, c(0.25, 0.5, 0.25))
return(apply(x == 1, 2, which) - 1)
}
generate_data <- function(n_indiv = 100,
n_loci_total = 20,
effect = c(1, -1, 0.5, -0.5, 0.25),
s_dev = 0.2)
{
pheno <- numeric(n_indiv)
geno <- matrix(nrow = n_indiv, ncol = n_loci_total)
for (i in 1:n_indiv) {
for (j in 1:n_loci_total) {
x <- generate_genotypes(n_loci_total)
}
geno[i, ] <- x
pheno[i] <- 0.0
n_loci_assoc <- length(effect)
for (j in 1:n_loci_assoc) {
pheno[i] <- pheno[i] + rnorm(1, effect[j]*geno[i,j], s_dev)
}
}
dat <- cbind(pheno, geno)
colnames(dat) <- c("pheno", paste("locus_", 1:n_loci_total, sep = ""))
return(as.data.frame(dat))
}
dat <- generate_data()
loci <- colnames(dat)[-1]
n_loci <- length(loci)
Running the analysis
We generated the data assuming that the individuals are all part of one, very large, well-mixed population. As a result, we don’t need to worry about relatedness as we did in Lab 13. We’ll use stan_lm()
for the analysis. It does a simple linear regression of phenotype on genotype, but as you can probably guess from the “stan” in its name, it uses Stan
as a backend to give is not only a posterior mean, but also credible intervals.
Locus-by-locus GWAS
We’ll start with a locus-by-locus GWAS like the one in Lab 13.
library(tidyverse)
Registered S3 methods overwritten by 'dbplyr':
method from
print.tbl_lazy
print.tbl_sql
── Attaching packages ───────────────────────────── tidyverse 1.3.1 ──
✓ ggplot2 3.3.5 ✓ purrr 0.3.4
✓ tibble 3.1.6 ✓ dplyr 1.0.7
✓ tidyr 1.1.4 ✓ stringr 1.4.0
✓ readr 2.1.0 ✓ forcats 0.5.1
── Conflicts ──────────────────────────────── tidyverse_conflicts() ──
x dplyr::filter() masks stats::filter()
x dplyr::lag() masks stats::lag()
library(rstanarm)
Loading required package: Rcpp
Registered S3 methods overwritten by 'htmltools':
method from
print.html tools:rstudio
print.shiny.tag tools:rstudio
print.shiny.tag.list tools:rstudio
Registered S3 method overwritten by 'htmlwidgets':
method from
print.htmlwidget tools:rstudio
This is rstanarm version 2.21.1
- See https://mc-stan.org/rstanarm/articles/priors for changes to default priors!
- Default priors may change, so it's safest to specify priors, even if equivalent to the defaults.
- For execution on a local, multicore CPU with excess RAM we recommend calling
options(mc.cores = parallel::detectCores())
options(mc.cores = parallel::detectCores())
results <- tibble(Locus = NA,
Mean = NA,
`2.5%` = NA,
`97.5%` = NA)
for (locus in 1:n_loci) {
fit <- stan_lm(paste("pheno ~ ", loci[locus], sep=""),
prior = R2(0.5, what = "mean"),
data = dat,
refresh = 0)
tmp <- data.frame(fit)
effect <- tmp[[loci[locus]]]
conf <- quantile(effect, c(0.025, 0.975))
results <- add_row(results,
Locus = locus,
Mean = round(mean(effect), 3),
`2.5%` = round(conf[1], 3),
`97.5%` = round(conf[2], 3))
}
results %>%
filter(!is.na(Mean)) %>%
arrange(desc(abs(Mean)))
Notice that locus 5, which we know has an allelic effect of 0.25 doesn’t appear among the top 5 in this list. In fact, it has the smallest estimated effect. It comes in at number 20. Similarly, locus 15, which we know doesn’t have an effect, appears in the top 5.
Genomic prediction
Genomic prediction is very similar to what we’ve just seen. We simply do one multiple regression in which all of the genotypes are included rather than doing a series of regressions separately for each locus. We start by constructing the regression_formula
, which looks pretty strange. We wouldn’t have to do this, but it’s easier than constructing the multiple regression formula by typing every locus into the formula.
regression_formula <- paste("pheno ~ ",
paste("locus_", 1:n_loci, sep = "",
collapse = " + "))
regression_formula
[1] "pheno ~ locus_1 + locus_2 + locus_3 + locus_4 + locus_5 + locus_6 + locus_7 + locus_8 + locus_9 + locus_10 + locus_11 + locus_12 + locus_13 + locus_14 + locus_15 + locus_16 + locus_17 + locus_18 + locus_19 + locus_20"
Now we’re ready to run the analysis and display the results. We’re going to use stan_glm()
with a gaussian()
family instead of stan_lm()
, because we want to use what’s known as a “shrinkage prior”. That’s a prior with the interesting property that either a covariate is included in the regression and the prior is fairly vague or it’s not included and the prior is tightly constrained around zero. It’s a way of letting the data tell us which covariates have strong associations and which don’t and simultaneously limiting the influence of those that don’t have strong associations on the results. But it does this without forcing us to pick particular covariates for the analysis.
The particular tool we use is what’s called a “horseshoe prior”. We only need to tell it one thing: How many coefficients we think might be important. The data will tell us how many actually are. p0
, which I set to 10 in this example is merely our best guess before we start the analysis.
n <- nrow(dat)
D <- ncol(dat) - 1
p0 <- 10
tau0 <- p0/(D - p0) * 1/sqrt(n)
prior_coeff <- hs(global_scale = tau0, slab_scale = 1)
fit <- stan_glm(as.formula(regression_formula),
family = gaussian(),
prior = prior_coeff,
data = dat,
refresh = 0)
fit_df <- as.data.frame(fit)
results_gp <- tibble(Locus = NA,
Mean = NA,
`2.5%` = NA,
`97.5%` = NA)
for (locus in 1:n_loci) {
tmp <- data.frame(fit)
effect <- tmp[[loci[locus]]]
conf <- quantile(effect, c(0.025, 0.975))
results_gp <- add_row(results_gp,
Locus = locus,
Mean = round(mean(effect), 3),
`2.5%` = round(conf[1], 3),
`97.5%` = round(conf[2], 3))
}
results_gp %>%
filter(!is.na(Mean)) %>%
arrange(desc(abs(Mean)))
Notice that with the genomic prediction approach we get the estimated allelic effects in the right order and roughly right in magnitude. This is only one example, and it is dangerous to extrapolate from one example, but if you’re familiar with multiple regression and its advantages, you’re probably wondering, “Why didn’t we just go directly with genomic prediction in Lab 13?”
Well, look at those data again. Even after severe filtering to remove loci that weren’t scored in at least 100 individuals, keeping only individuals scored at more than 4000 of the remaining loci, and excluding any loci where one or more of the remaining individuals wasn’t scored we had data from 218 loci and 141 individuals. That means we have 141 observations that we are trying to “predict” from 218 covariates. We have more covariates than observations, and for multiple linear regression to work we need more observations than covariates. To get good estimates of regression coefficients you need several observations per coefficient. Here we did pretty well with 5 observations per coefficient. You might want to try increasing the number of loci included in the data set to 50, keeping effect
set at the default and tyring the simulation again to see what you get.
Comparing one-locus GWAS and genomic predictions
Let’s first look at the locus-by-locus estimates of allelic effects.
for_plot <- tibble(GWAS = results$Mean,
GP = results_gp$Mean) %>%
filter(!is.na(GWAS))
p <- ggplot(for_plot, aes(x = GWAS, y = GP)) +
geom_point() +
geom_abline(slope = 1, intercept = 0, color = "red",
linetype = "dashed") +
theme_bw()
p
![](data:image/png;base64,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)
They are broadly similar, but there are also clearly some differences. Now let’s see how well we can predict the phenotypes.
predictions <- tibble(Observed = NA,
Model = NA,
Predicted = NA,
Error = NA)
for (i in 1:nrow(dat)) {
predicted <-0.0
for (j in 1:n_loci) {
predicted <- predicted + results$Mean[i]*dat[i, j+1]
}
predictions <- add_row(predictions,
Observed = dat$pheno[i],
Model = "GWAS",
Predicted = predicted,
Error = Predicted - Observed)
predicted <-0.0
for (j in 1:n_loci) {
predicted <- predicted + results_gp$Mean[i]*dat[i, j+1]
}
predictions <- add_row(predictions,
Observed = dat$pheno[i],
Model = "GP",
Predicted = predicted,
Error = Predicted - Observed)
}
predictions <- filter(predictions, !is.na(Predicted))
p <- ggplot(predictions, aes(x = Observed, y = Predicted)) +
geom_point() +
geom_abline(slope = 1, intercept = 0, color = "red",
linetype = "dashed") +
facet_wrap(~ Model) +
theme_bw()
p
![](data:image/png;base64,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)
You can see that there’s a cluster of points close to the 1:1 line with genomic prediction, and that the points seem to be a bit farther from the 1:1 line with the locus by locus GWAS. We can check that by calculating the mean squared error.
cat("Mean squared error:\n",
" GWAS: ", round(sum(subset(predictions, Model == "GWAS")$Error^2),
3)/nrow(dat), "\n",
" GP: ", round(sum(subset(predictions, Model == "GP")$Error^2),
3)/nrow(dat), "\n",
sep = "")
Mean squared error:
GWAS: 6.01941
GP: 8.02203
As you can see, our (or at least my) subjective impression was wrong. If we were to use all 20 loci, the locus by locus GWAS would, on average, be closer to the “truth” than the genomic prediction approach – even though we know that it got the effects wrong.
---
title: "Exploring genomic prediction"
output: 
  html_notebook:
    toc: yes
    toc_float: true
    css: "R-notebook.css"
---

# Overview

This `R` notebook may be a simpler way to follow along with the lecture notes that introduce the idea of genomic prediction ([HTML](http://darwin.eeb.uconn.edu/eeb348-notes/genomic-prediction.html), [PDF](http://darwin.eeb.uconn.edu/eeb348-notes/genomic-prediction.pdf)). Much of the code is taken directly from [http://darwin.eeb.uconn.edu/eeb348-resources/genomic-prediction.R](http://darwin.eeb.uconn.edu/eeb348-resources/genomic-prediction.R) with some minor modifications to reflect things I've learned about coding in `R` in the last two and a half years. The final part ([Comparing one-locus GWAS and genomic predictions](#comparing-one-locus-gwas-and-genomic-predictions)) includes some new code.

# Generating the data for analysis

The first step in exploring genomic prediction is to generate data where we know the answer so that we can compare our estimate from the data with the truth. `generate_data()` does this by following these steps.

1. Generate an genotype at each locus for an individual. By default, `generate_data()` generates genotypes at 20 loci, but you can change that when you call it by changing `n_loci_total` in the call. The genotype at each locus is generated randomly assuming that the genotypes are in proportions 0.25, 0.5, and 0.25 for one homozygote, the heterozygote, and the other homozygote respectively.

2. Generate a phenotype for this individual.

    a. Calculate the genotypic mean at each locus from the specified allelic effect and the genotype at that locus. By default the allelic effects are 1.0, -1.0, 0.5, -0.5, and 0.25 at loci 1-5 and 0.0 at the remaining 15 loci. You can change that by changing `effect` when you call `generate_data()`.
    b. Generate a phenotypic contribution each locus as a random variable having a normal distribution with the genotypic mean and a specified standard deviation. By default the standard deviation is 0.2. You can change that by changing `s_dev` when you call `generate_data()`.
    c. Sum the phenotypic contribution across all of the loci.^[Since we're assuming that only loci 1-5 influence the phenotype, we only do this sum across loci 1-5.]
  
3. Repeat #1 and #2 until you have the number of individuals in the sample data set that you want. By default that's 100, but you can change that by changing `n_indiv` when you call `generate_data`.

`generate_data()` returns the results of the simulation as a data frame with each individual on a row and with the phenotype in the first column and the genotypes at each locus in the remaining columns.

NOTE: I'm using `set.seed()` to ensure that you get the same sequence of random numbers that I do if you run this code on your own. You can delete that line or comment it out if you want to see what happens with different randomly generated data sets.

```{r}
set.seed(12345)

generate_genotypes <- function(n) {
  x <- rmultinom(n, 1, c(0.25, 0.5, 0.25))
  return(apply(x == 1, 2, which) - 1)
}

generate_data <- function(n_indiv = 100,
                          n_loci_total = 20,
                          effect = c(1, -1, 0.5, -0.5, 0.25),
                          s_dev = 0.2)
{
  pheno <- numeric(n_indiv)
  geno <- matrix(nrow = n_indiv, ncol = n_loci_total)
  for (i in 1:n_indiv) {
    for (j in 1:n_loci_total) {
      x <- generate_genotypes(n_loci_total)
    }
    geno[i, ] <- x
    pheno[i] <- 0.0
    n_loci_assoc <- length(effect)
    for (j in 1:n_loci_assoc) {
      pheno[i] <- pheno[i] + rnorm(1, effect[j]*geno[i,j], s_dev)
    }
  }
  dat <- cbind(pheno, geno)
  colnames(dat) <- c("pheno", paste("locus_", 1:n_loci_total, sep = ""))
  return(as.data.frame(dat))
}

dat <- generate_data()
loci <- colnames(dat)[-1]
n_loci <- length(loci)
```

# Running the analysis

We generated the data assuming that the individuals are all part of one, very large, well-mixed population. As a result, we don't need to worry about relatedness as we did in [Lab 13](http://darwin.eeb.uconn.edu/eeb348-resources/Lab13.nb.html). We'll use `stan_lm()` for the analysis. It does a simple linear regression of phenotype on genotype, but as you can probably guess from the "stan" in its name, it uses `Stan` as a backend to give is not only a posterior mean, but also credible intervals.

## Locus-by-locus GWAS

We'll start with a locus-by-locus GWAS like the one in [Lab 13](http://darwin.eeb.uconn.edu/eeb348-resources/Lab13.nb.html). 

```{r}
library(tidyverse)
library(rstanarm)

options(mc.cores = parallel::detectCores())

results <- tibble(Locus = NA,
                  Mean = NA,
                  `2.5%` = NA,
                  `97.5%` = NA)
for (locus in 1:n_loci) {
  fit <- stan_lm(paste("pheno ~ ", loci[locus], sep=""),
                 prior = R2(0.5, what = "mean"),
                 data = dat,
                 refresh = 0)
  tmp <- data.frame(fit)
  effect <- tmp[[loci[locus]]]
  conf <- quantile(effect, c(0.025, 0.975))
  results <- add_row(results,
                     Locus = locus,
                     Mean = round(mean(effect), 3),
                     `2.5%` = round(conf[1], 3),
                     `97.5%` = round(conf[2], 3))
}
results %>%
  filter(!is.na(Mean)) %>%
  arrange(desc(abs(Mean)))
```

Notice that locus 5, which we know has an allelic effect of 0.25 doesn't appear among the top 5 in this list. In fact, it has the *smallest* estimated effect. It comes in at number 20. Similarly, locus 15, which we know doesn't have an effect, appears in the top 5.

## Genomic prediction

Genomic prediction is very similar to what we've just seen. We simply do one multiple regression in which all of the genotypes are included rather than doing a series of regressions separately for each locus. We start by constructing the `regression_formula`, which looks pretty strange. We wouldn't have to do this, but it's easier than constructing the multiple regression formula by typing every locus into the formula.

```{r}
regression_formula <- paste("pheno ~ ", 
                            paste("locus_", 1:n_loci, sep = "",
                                  collapse = " + "))
regression_formula
```
Now we're ready to run the analysis and display the results. We're going to use `stan_glm()` with a `gaussian()` family instead of `stan_lm()`, because we want to use what's known as a "shrinkage prior". That's a prior with the interesting property that either a covariate is included in the regression and the prior is fairly vague or it's not included and the prior is tightly constrained around zero. It's a way of letting the data tell us which covariates have strong associations and which don't and simultaneously limiting the influence of those that don't have strong associations on the results. But it does this without forcing us to pick particular covariates for the analysis.

The particular tool we use is what's called a "horseshoe prior". We only need to tell it one thing: How many coefficients we think might be important. The data will tell us how many actually are. `p0`, which I set to 10 in this example is merely our best guess before we start the analysis.

```{r}
n <- nrow(dat)
D <- ncol(dat) - 1
p0 <- 10
tau0 <- p0/(D - p0) * 1/sqrt(n)
prior_coeff <- hs(global_scale = tau0, slab_scale = 1)

fit <- stan_glm(as.formula(regression_formula),
                family = gaussian(),
                prior = prior_coeff,
                data = dat,
                refresh = 0)
fit_df <- as.data.frame(fit)

results_gp <- tibble(Locus = NA,
                     Mean = NA,
                     `2.5%` = NA,
                     `97.5%` = NA)
for (locus in 1:n_loci) {
  tmp <- data.frame(fit)
  effect <- tmp[[loci[locus]]]
  conf <- quantile(effect, c(0.025, 0.975))
  results_gp <- add_row(results_gp,
                        Locus = locus,
                        Mean = round(mean(effect), 3),
                       `2.5%` = round(conf[1], 3),
                       `97.5%` = round(conf[2], 3))
}
results_gp %>%
  filter(!is.na(Mean)) %>%
  arrange(desc(abs(Mean)))
```

Notice that with the genomic prediction approach we get the estimated allelic effects in the right order and roughly right in magnitude. This is only one example, and it is dangerous to extrapolate from one example, but if you're familiar with multiple regression and its advantages, you're probably wondering, "Why didn't we just go directly with genomic prediction in [Lab 13](http://darwin.eeb.uconn.edu/eeb348-resources/Lab13.nb.html)?" 

Well, look at those data again. Even after severe filtering to remove loci that weren't scored in at least 100 individuals, keeping only individuals scored at more than 4000 of the remaining loci, and excluding any loci where one or more of the remaining individuals wasn't scored we had data from 218 loci and 141 individuals. That means we have 141 observations that we are trying to "predict" from 218 covariates. We have more covariates than observations, and for multiple linear regression to work we need more observations than covariates. To get good estimates of regression coefficients you need several observations per coefficient. Here we did pretty well with 5 observations per coefficient. You might want to try increasing the number of loci included in the data set to 50, keeping `effect` set at the default and tyring the simulation again to see what you get.

# Comparing one-locus GWAS and genomic predictions

Let's first look at the locus-by-locus estimates of allelic effects.

```{r}
for_plot <- tibble(GWAS = results$Mean,
                   GP = results_gp$Mean) %>%
  filter(!is.na(GWAS))
p <- ggplot(for_plot, aes(x = GWAS, y = GP)) +
  geom_point() +
  geom_abline(slope = 1, intercept = 0, color = "red",
              linetype = "dashed") +
  theme_bw()
p
```
They are broadly similar, but there are also clearly some differences. Now let's see how well we can predict the phenotypes.

```{r}
predictions <- tibble(Observed = NA,
                      Model = NA,
                      Predicted = NA,
                      Error = NA)
for (i in 1:nrow(dat)) {
  predicted <-0.0
  for (j in 1:n_loci) {
    predicted <- predicted + results$Mean[i]*dat[i, j+1]
  }
  predictions <- add_row(predictions,
                         Observed = dat$pheno[i],
                         Model = "GWAS",
                         Predicted = predicted,
                         Error = Predicted - Observed)
  predicted <-0.0
  for (j in 1:n_loci) {
    predicted <- predicted + results_gp$Mean[i]*dat[i, j+1]
  }
  predictions <- add_row(predictions,
                         Observed = dat$pheno[i],
                         Model = "GP",
                         Predicted = predicted,
                         Error = Predicted - Observed)
}
predictions <- filter(predictions, !is.na(Predicted))
p <- ggplot(predictions, aes(x = Observed, y = Predicted)) +
  geom_point() +
  geom_abline(slope = 1, intercept = 0, color = "red", 
              linetype = "dashed") +
  facet_wrap(~ Model) +
  theme_bw()
p
```

You can see that there's a cluster of points close to the 1:1 line with genomic prediction, and that the points seem to be a bit farther from the 1:1 line with the locus by locus GWAS. We can check that by calculating the mean squared error.

```{r}
cat("Mean squared error:\n",
    "   GWAS: ", round(sum(subset(predictions, Model == "GWAS")$Error^2),
                       3)/nrow(dat), "\n",
    "     GP: ", round(sum(subset(predictions, Model == "GP")$Error^2),
                       3)/nrow(dat), "\n",
    sep = "")
```

As you can see, our (or at least my) subjective impression was wrong. If we were to use all 20 loci, the locus by locus GWAS would, on average, be closer to the "truth" than the genomic prediction approach -- even though we know that it got the effects wrong.