hm450.manifest.pop.GoNL: HM450 population-specific probe-masking recommendations

Description Usage Format Details Source References Examples

Description

Adapted version of the annotation file provided by Zhou et al. (see source, Mar-13-2017 release). This annotation file contains population-specific probe-masking recommendations based on SNPs within 5 bases from the 3'end of the probe, mapping issues, non-unique 3' 30bp subsequence and channel-switching SNPs in the single-base-extension for type I probes. We added population-specific masking recommendations for the Dutch population using GoNL release 5. This release does not include X and Y chromosomes, so for the Dutch population, only masking information for the autosomal probes is available.

Usage

1

Format

A GRanges object with 485577 ranges and 65 metadata columns:

MASK.general.pop

Recommended general purpose masking merged from "MASK.sub30.copy", "MASK.mapping" (in either the hg38 or hg19 genome), "MASK.extBase", "MASK.typeINextBaseSwitch" and "MASK.snp5.pop" from the "hm450.manifest" file and the "hm450.manifest.pop" file (see source).For GoNL, "MASK.typeINextBaseSwitchandINDEL.GoNL" is used instead of "MASK.typeINextBaseSwitch"

MASK.snp5.pop

Whether the 5bp 3'-subsequence (including extension for type II) overlap with a SNP with population-specific AF > 0.01

MASK.typeINextBaseSwitchandINDEL.GoNL

SNPs (that cause a color-channel switch) and INDELS with AF > 0.01 in GoNL. In contrast, "MASK.typeINextBaseSwitch" column is based on all SNPs in 1000 genomes and dbSNP, regardless of population or allele frequency

Details

Note: Zhou et al. identified several probes that match to a different location than annotated in the original Illumina manifest file. The authors have used the 'updated' location in their annotation file. Therefore, a handful of probes in this annotation file are annotated to a different location than the original Illumina manifest file. For the identification of the overlaps we used the locations as annotated in the original Illumina file. Therefore, a few of the identified overlaps do not match the locations as specified in this annotation file. This also explains why GoNL masking information is available for a couple of probes that are located in the X- and Y-chromosome in this annotation: these probes map to autosomal probes in the original Illumina file. All probes that map to a different location than originally annotated are recommended to be masked (in the MASK.mapping and MASK.general column), so generally they won't be included in further analyses.

Source

http://zwdzwd.github.io/InfiniumAnnotation

References

Zhou W, Laird PW and Shen H: Comprehensive characterization, annotation and innovative use of Infinium DNA Methylation BeadChip probes. Nucleic Acids Research 2016

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
# Select probes that should be masked in Dutch population 
# (note that X and Y chromosomes are not included)
hm450.manifest.pop.GoNL <- hm450.manifest.pop.GoNL[!is.na(
    hm450.manifest.pop.GoNL$MASK.general.GoNL) &
    hm450.manifest.pop.GoNL$MASK.general.GoNL == TRUE, ]
  
# Select probes that should be masked in Dutch population because there is 
# a SNP within 5 bases of the 3'end of the probe 
# (note that X and Y chromosomes are not included)
hm450.manifest.pop.GoNL <- hm450.manifest.pop.GoNL[!is.na(
    hm450.manifest.pop.GoNL$MASK.snp5.GoNL) &
    hm450.manifest.pop.GoNL$MASK.snp5.GoNL == TRUE, ]
    
# When studying a Dutch population and one wants to include X and Y 
# chromosomal probes, the EUR or CEU population can be used.                                                      
# Select probes that should be masked in European population 
# (these include X and Y chromosomes)                                                      
hm450.manifest.pop.GoNL <- hm450.manifest.pop.GoNL[
    hm450.manifest.pop.GoNL$MASK.general.EUR == TRUE,]

omicsPrint documentation built on Nov. 8, 2020, 4:55 p.m.