| Title: | Identify and Rank CpG DNA Methylation Conservation Along the Human Genome |
| Version: | 0.1.0 |
| Description: | Identify and rank CpG DNA methylation conservation along the human genome. Specifically it includes bootstrapping methods to provide ranking which should adjust for the differences in length as without it short regions tend to get higher conservation scores. |
| License: | MIT + file LICENSE |
| Encoding: | UTF-8 |
| LazyData: | true |
| RoxygenNote: | 7.0.2 |
| Depends: | R (≥ 2.10) |
| Imports: | magrittr, dplyr, purrr, rlang |
| Suggests: | ggplot2, testthat (≥ 2.1.0), covr |
| URL: | https://github.com/EmilHvitfeldt/methcon5 |
| BugReports: | https://github.com/EmilHvitfeldt/methcon5/issues |
| NeedsCompilation: | no |
| Packaged: | 2019-12-17 02:47:31 UTC; emilhvitfeldthansen |
| Author: | Emil Hvitfeldt 
[aut, cre] |
| Maintainer: | Emil Hvitfeldt <emilhhvitfeldt@gmail.com> |
| Repository: | CRAN |
| Date/Publication: | 2019-12-20 13:50:02 UTC |
Pipe operator
Description
See magrittr::%>% for details.
Usage
lhs %>% rhs
Simple simulated methylation dataset
Description
Simple simulated methylation dataset
Usage
fake_methylation
Format
A data frame with 2771 rows and 3 variables: gene,
cons_level and meth.
Details
This dataset is for example use only. It contains 500 genes
identified by gene each with one of 3 types of conservation levels
"low", "medium" and "high". The methylation values are independent randomly
distributed within each gene. Thus no spacial correlation is assumed.
Calculate region wise summary statistics
Description
Will take a data.frame and apply a function ('fun') to 'value' within the groups defined by the 'id' column.
Usage
meth_aggregate(data, id, value, fun = mean, ...)
Arguments
data |
a data.frame. |
id |
variable name, to be aggregated around. |
value |
variable name, contains the value to take mean over. Must be a single column. |
fun |
function, summary statistic function to be calculated. Defaults to 'mean'. |
... |
Additional arguments for the function given to the argument fun. |
Details
Please note the ordering of the data will matter depending on the choice of aggregation function.
Value
A methcon object. Contains the aggregated data along with original data.frame and variable selections.
Examples
meth_aggregate(fake_methylation, id = gene, value = meth, fun = mean)
meth_aggregate(fake_methylation, id = gene, value = meth, fun = var)
# custom functions can be used as well
mean_diff <- function(x) {
mean(diff(x))
}
meth_aggregate(fake_methylation, id = gene, value = meth, fun = mean_diff)
Bootstrapped randomly samples values
Description
"perm_v1" (the default method) will sample the variables the rows independently. "perm_v2" will sample regions of same size while allowing overlap between different regions. "perm_v3" will sample regions under the constraint that all sampled regions are contained in the region they are sampled in.
Usage
meth_bootstrap(data, reps, method = c("perm_v1", "perm_v2", "perm_v3"))
Arguments
data |
a methcon data.frame output from 'meth_bootstrap'. |
reps |
Number of reps, defaults to 1000. |
method |
Character, determining which method to use. See details for information about methods. Defaults to "perm_v1". |
Details
Note that you can apply 'meth_bootstrap' multiple times to get values for different methods.
Value
A methcon object. Contains the aggregated data along with original data.frame and variable selections and bootstrapped values.
Examples
# Note that you likely want to do more than 10 repitions.
# rep = 10 was chosen to have the examples run fast.
fake_methylation %>%
meth_aggregate(id = gene, value = meth, fun = mean) %>%
meth_bootstrap(10)
fake_methylation %>%
meth_aggregate(id = gene, value = meth, fun = mean) %>%
meth_bootstrap(10, method = "perm_v2")
# Get multiple bootstraps
fake_methylation %>%
meth_aggregate(id = gene, value = meth, fun = mean) %>%
meth_bootstrap(10, method = "perm_v1") %>%
meth_bootstrap(10, method = "perm_v2") %>%
meth_bootstrap(10, method = "perm_v3")