Correlates of inverse language ability and methylation - compartment analysis
Mark Ziemann
2023-03-20
Source: https://github.com/markziemann/asd_meth
Introduction
Here I will be running a comparison of differential methylation data from guthrie and fresh blood samples.
Here are the files that I’m using:
limma_guthrie_ilanguage.csv
limma_blood_ilanguage.csv
suppressPackageStartupMessages({
library("parallel")
library("mitch")
library("IlluminaHumanMethylationEPICanno.ilm10b4.hg19")
source("https://raw.githubusercontent.com/markziemann/gmea/main/meth_functions.R")
library("data.table")
library("kableExtra")
library("eulerr")
library("RIdeogram")
library("GenomicRanges")
library("tictoc")
})
source("meth_functions.R")Load Limma and RUV data for manhattan plot
First I will generate plots for limmma analysis.
# blood at assessment
top <- read.csv("limma_blood_ilanguage.csv")
nrow(top)## [1] 802647top <- subset(top,P.Value<1e-2)
nrow(top)## [1] 164412-log10(min(top$P.Value))## [1] 7.249305top$chr <- as.integer(gsub("chr","",top$chr))
top$snp <- paste(top$Name,top$UCSC_RefGene_Name)
top$snp <- sapply(strsplit(top$snp,";"),"[[",1)
up <- subset(top,logFC>0)
dn <- subset(top,logFC<0)
par(mfrow=c(2,1))
par(mar=c(4,4,3,5))
manhattan(x=up,chr="chr",bp="pos",p="P.Value",snp="snp",suggestiveline = -log10(1e-04),
ylim=c(2,6), annotatePval= 1e-04 ,main="Blood at assessment limma hypermethylated")
manhattan(x=dn,chr="chr",bp="pos",p="P.Value",snp="snp",suggestiveline = -log10(1e-04),
ylim=c(2,6), annotatePval= 1e-04 , main="Blood at assessment limma hypomethylated")
pdf("manhat_limma_bl.pdf")
par(mfrow=c(2,1))
par(mar=c(4,4,3,5))
manhattan(x=up,chr="chr",bp="pos",p="P.Value",snp="snp",suggestiveline = -log10(1e-04),
ylim=c(2,6), annotatePval= 1e-04 ,main="Blood at assessment limma hypermethylated")
manhattan(x=dn,chr="chr",bp="pos",p="P.Value",snp="snp",suggestiveline = -log10(1e-04),
ylim=c(2,6), annotatePval= 1e-04 , main="Blood at assessment limma hypomethylated")
dev.off()## png
## 2# neonatal guthrie card
top <- read.csv("limma_guthrie_ilanguage.csv")
nrow(top)## [1] 790658top <- subset(top,P.Value<1e-2)
nrow(top)## [1] 5255-log10(min(top$P.Value))## [1] 7.038618top$chr <- as.integer(gsub("chr","",top$chr))
top$snp <- paste(top$Name,top$UCSC_RefGene_Name)
top$snp <- sapply(strsplit(top$snp,";"),"[[",1)
up <- subset(top,logFC>0)
dn <- subset(top,logFC<0)
par(mfrow=c(2,1))
par(mar=c(4,4,3,5))
manhattan(x=up,chr="chr",bp="pos",p="P.Value",snp="snp",suggestiveline = -log10(1e-04),
ylim=c(2,6), annotatePval= 1e-04 ,main="Neonatal Guthrie card limma hypermethylated",
annotateTop=FALSE)
manhattan(x=dn,chr="chr",bp="pos",p="P.Value",snp="snp",suggestiveline = -log10(1e-04),
ylim=c(2,6), annotatePval= 1e-04 , main="Neonatal Guthrie card limma hypomethylated",
annotateTop=FALSE)
pdf("manhat_limma_gu.pdf")
par(mfrow=c(2,1))
par(mar=c(4,4,3,5))
manhattan(x=up,chr="chr",bp="pos",p="P.Value",snp="snp",suggestiveline = -log10(1e-04),
ylim=c(2,6), annotatePval= 1e-04 ,main="Neonatal Guthrie card limma hypermethylated",
annotateTop=FALSE)
manhattan(x=dn,chr="chr",bp="pos",p="P.Value",snp="snp",suggestiveline = -log10(1e-04),
ylim=c(2,6), annotatePval= 1e-04 , main="Neonatal Guthrie card limma hypomethylated",
annotateTop=FALSE)
dev.off()## png
## 2Compartment enrichment
Looking for enrichments in different genomic compartments.
Guthrie cards.
par(mfrow=c(2,1))
# guthrie
dma1 <- read.csv("limma_guthrie_ilanguage.csv")
if (nrow(subset(dma1,P.Value <1e-4)) > 100 ) {
comp <- compartment_enrichment(dma1)
comp <- lapply(comp,function(x) {subset(x,OR!=0)} )
comp
make_forest_plots_up(comp)
make_forest_plots_dn(comp)
}Blood at assessment.
# blood
dma2 <- read.csv("limma_blood_ilanguage.csv")
if (nrow(subset(dma2,P.Value <1e-4)) > 100 ) {
comp <- compartment_enrichment(dma2)
comp <- lapply(comp,function(x) {subset(x,OR!=0)} )
comp
make_forest_plots_up(comp)
make_forest_plots_dn(comp)
}
Compartments top 1000
Guthrie cards.
par(mfrow=c(2,1))
# guthrie
comp <- compartment_enrichment2(dma1)
comp <- lapply(comp,function(x) {subset(x,OR!=0)} )
comp## $up_comp
## all up OR fisherPval lowerCI upperCI
## Intergenic 43381 94 1.5329042 2.023156e-04 1.2261550 1.8972837
## 3'UTR 22433 19 0.5713113 1.247369e-02 0.3424494 0.8971594
## 5'UTR 102937 175 1.1975147 3.382504e-02 1.0116677 1.4109891
## Body 333689 398 0.6988926 1.905590e-08 0.6147317 0.7938427
## ExonBnd 6713 6 0.6090342 2.620135e-01 0.2228945 1.3309217
## TSS1500 115845 193 1.1730093 4.781774e-02 0.9975967 1.3737349
## TSS200 75104 138 1.2981778 5.379668e-03 1.0768669 1.5551207
##
## $dn_comp
## all dn OR fisherPval lowerCI upperCI
## Intergenic 43381 55 1.0129311 0.888593474 0.7563252 1.3319095
## 3'UTR 22433 34 1.2187059 0.249109481 0.8381156 1.7171137
## 5'UTR 102937 125 0.9642591 0.738526390 0.7913917 1.1667716
## Body 333689 444 1.1261296 0.084448527 0.9841760 1.2886336
## ExonBnd 6713 11 1.3125128 0.380863357 0.6524324 2.3617552
## TSS1500 115845 145 0.9990403 1.000000000 0.8300653 1.1955092
## TSS200 75104 63 0.6437697 0.000459275 0.4899999 0.8325786make_forest_plots_up(comp)
make_forest_plots_dn(comp)
Blood at assessment.
# blood
comp <- compartment_enrichment2(dma2)
comp <- lapply(comp,function(x) {subset(x,OR!=0)} )
comp## $up_comp
## all up OR fisherPval lowerCI upperCI
## Intergenic 43528 12 0.2892024 2.698376e-07 0.1486264 0.5087691
## 3'UTR 22758 10 0.4739116 1.343508e-02 0.2261303 0.8774397
## 5'UTR 104442 90 0.9369575 6.171752e-01 0.7413338 1.1726210
## Body 339559 395 1.7113584 1.687712e-11 1.4574293 2.0125828
## ExonBnd 6862 7 1.1210534 6.855946e-01 0.4487353 2.3244830
## TSS1500 117509 103 0.9547975 7.110641e-01 0.7657451 1.1807531
## TSS200 75457 30 0.4087208 6.475086e-08 0.2733558 0.5895656
##
## $dn_comp
## all dn OR fisherPval lowerCI upperCI
## Intergenic 43528 25 0.4417693 6.772985e-06 0.2842532 0.6567026
## 3'UTR 22758 39 1.3831626 5.598552e-02 0.9763397 1.9072498
## 5'UTR 104442 126 0.9550911 6.701916e-01 0.7846131 1.1546695
## Body 339559 518 1.5162874 7.252598e-10 1.3246448 1.7369773
## ExonBnd 6862 10 1.1635181 6.042388e-01 0.5557401 2.1512650
## TSS1500 117509 136 0.9083421 3.210465e-01 0.7510066 1.0917089
## TSS200 75457 37 0.3640932 2.998699e-12 0.2545800 0.5059732make_forest_plots_up(comp)
make_forest_plots_dn(comp)
Looking for enrichments in proximity to CGI contexts
Guthrie cards.
par(mfrow=c(2,1))
if (nrow(subset(dma1,P.Value <1e-4)) > 100 ) {
# guthrie
comp <- cgi_enrichment(dma1)
comp <- lapply(comp,function(x) {subset(x,OR!=0)} )
comp
make_forest_plots_up(comp)
make_forest_plots_dn(comp)
}Blood at assessment.
if (nrow(subset(dma2,P.Value <1e-4)) > 100 ) {
# blood
comp <- cgi_enrichment(dma2)
comp <- lapply(comp,function(x) {subset(x,OR!=0)} )
comp
make_forest_plots_up(comp)
make_forest_plots_dn(comp)
}
CGI enrichments in top 1000 probes
Guthrie cards.
par(mfrow=c(2,1))
# guthrie
comp <- cgi_enrichment2(dma1)
comp <- lapply(comp,function(x) {subset(x,OR!=0)} )
comp## $up_comp
## all up OR fisherPval lowerCI upperCI
## Island 150182 326 2.0650990 1.697933e-24 1.8034688 2.3606663
## OpenSea 442427 378 0.4778792 1.224277e-30 0.4193866 0.5439918
## Shelf 55491 68 0.9665802 8.525393e-01 0.7441240 1.2374634
## Shore 142558 228 1.3432132 1.485704e-04 1.1534664 1.5592217
##
## $dn_comp
## all dn OR fisherPval lowerCI upperCI
## Island 150182 178 0.9234262 0.3535806069 0.7807329 1.0871899
## OpenSea 442427 518 0.8457016 0.0089327487 0.7455330 0.9594641
## Shelf 55491 80 1.1522533 0.2156910627 0.9050202 1.4493926
## Shore 142558 224 1.3128010 0.0004602369 1.1262334 1.5253509make_forest_plots_up(comp)
make_forest_plots_dn(comp)
Blood at assessment.
# blood
comp <- cgi_enrichment2(dma2)
comp <- lapply(comp,function(x) {subset(x,OR!=0)} )
comp## $up_comp
## all up OR fisherPval lowerCI upperCI
## Island 150186 14 0.06159713 2.175707e-68 0.03357243 0.1038500
## OpenSea 452360 853 4.49991341 5.942087e-86 3.77308568 5.3976214
## Shelf 56174 50 0.69911498 1.290409e-02 0.51512263 0.9295346
## Shore 143927 83 0.41391359 4.883433e-18 0.32652705 0.5186457
##
## $dn_comp
## all dn OR fisherPval lowerCI upperCI
## Island 150186 39 0.1760916 6.262393e-45 0.1244179 0.2425214
## OpenSea 452360 752 2.3502472 3.513057e-35 2.0332729 2.7242559
## Shelf 56174 68 0.9695163 8.525011e-01 0.7463704 1.2412212
## Shore 143927 141 0.7510079 1.283048e-03 0.6238892 0.8984489make_forest_plots_up(comp)
make_forest_plots_dn(comp)
Session information
For reproducibility
sessionInfo()## R version 4.2.2 Patched (2022-11-10 r83330)
## Platform: x86_64-pc-linux-gnu (64-bit)
## Running under: Ubuntu 22.04.2 LTS
##
## Matrix products: default
## BLAS: /usr/lib/x86_64-linux-gnu/blas/libblas.so.3.10.0
## LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.10.0
##
## locale:
## [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
## [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
## [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
## [7] LC_PAPER=en_US.UTF-8 LC_NAME=C
## [9] LC_ADDRESS=C LC_TELEPHONE=C
## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
##
## attached base packages:
## [1] grid stats4 parallel stats graphics grDevices utils
## [8] datasets methods base
##
## other attached packages:
## [1] tictoc_1.1
## [2] RIdeogram_0.2.2
## [3] kableExtra_1.3.4
## [4] data.table_1.14.8
## [5] ENmix_1.34.0
## [6] doParallel_1.0.17
## [7] qqman_0.1.8
## [8] RCircos_1.2.2
## [9] beeswarm_0.4.0
## [10] forestplot_3.1.1
## [11] abind_1.4-5
## [12] checkmate_2.1.0
## [13] reshape2_1.4.4
## [14] gplots_3.1.3
## [15] eulerr_7.0.0
## [16] GEOquery_2.66.0
## [17] RColorBrewer_1.1-3
## [18] IlluminaHumanMethylation450kmanifest_0.4.0
## [19] topconfects_1.14.0
## [20] DMRcatedata_2.16.0
## [21] ExperimentHub_2.6.0
## [22] AnnotationHub_3.6.0
## [23] BiocFileCache_2.6.0
## [24] dbplyr_2.3.1
## [25] DMRcate_2.12.0
## [26] limma_3.54.0
## [27] missMethyl_1.32.0
## [28] IlluminaHumanMethylation450kanno.ilmn12.hg19_0.6.1
## [29] R.utils_2.12.2
## [30] R.oo_1.25.0
## [31] R.methodsS3_1.8.2
## [32] plyr_1.8.8
## [33] IlluminaHumanMethylationEPICanno.ilm10b4.hg19_0.6.0
## [34] minfi_1.44.0
## [35] bumphunter_1.40.0
## [36] locfit_1.5-9.7
## [37] iterators_1.0.14
## [38] foreach_1.5.2
## [39] Biostrings_2.66.0
## [40] XVector_0.38.0
## [41] SummarizedExperiment_1.28.0
## [42] Biobase_2.58.0
## [43] MatrixGenerics_1.10.0
## [44] matrixStats_0.63.0
## [45] GenomicRanges_1.50.2
## [46] GenomeInfoDb_1.34.6
## [47] IRanges_2.32.0
## [48] S4Vectors_0.36.1
## [49] BiocGenerics_0.44.0
## [50] mitch_1.10.0
##
## loaded via a namespace (and not attached):
## [1] rappdirs_0.3.3 rtracklayer_1.58.0
## [3] GGally_2.1.2 tidyr_1.3.0
## [5] ggplot2_3.4.1 bit64_4.0.5
## [7] knitr_1.42 DelayedArray_0.24.0
## [9] rpart_4.1.19 KEGGREST_1.38.0
## [11] RCurl_1.98-1.10 AnnotationFilter_1.22.0
## [13] generics_0.1.3 GenomicFeatures_1.50.3
## [15] preprocessCore_1.60.2 RSQLite_2.3.0
## [17] bit_4.0.5 tzdb_0.3.0
## [19] webshot_0.5.4 xml2_1.3.3
## [21] httpuv_1.6.9 assertthat_0.2.1
## [23] xfun_0.37 hms_1.1.2
## [25] jquerylib_0.1.4 evaluate_0.20
## [27] promises_1.2.0.1 fansi_1.0.4
## [29] restfulr_0.0.15 scrime_1.3.5
## [31] progress_1.2.2 caTools_1.18.2
## [33] readxl_1.4.2 DBI_1.1.3
## [35] geneplotter_1.76.0 htmlwidgets_1.6.2
## [37] reshape_0.8.9 purrr_1.0.1
## [39] ellipsis_0.3.2 dplyr_1.1.0
## [41] backports_1.4.1 permute_0.9-7
## [43] calibrate_1.7.7 grImport2_0.2-0
## [45] annotate_1.76.0 biomaRt_2.54.0
## [47] deldir_1.0-6 sparseMatrixStats_1.10.0
## [49] vctrs_0.6.0 ensembldb_2.22.0
## [51] withr_2.5.0 cachem_1.0.7
## [53] Gviz_1.42.0 BSgenome_1.66.2
## [55] GenomicAlignments_1.34.0 prettyunits_1.1.1
## [57] mclust_6.0.0 svglite_2.1.1
## [59] cluster_2.1.4 RPMM_1.25
## [61] lazyeval_0.2.2 crayon_1.5.2
## [63] genefilter_1.80.3 edgeR_3.40.2
## [65] pkgconfig_2.0.3 nlme_3.1-162
## [67] ProtGenerics_1.30.0 nnet_7.3-18
## [69] rlang_1.1.0 lifecycle_1.0.3
## [71] filelock_1.0.2 dichromat_2.0-0.1
## [73] rsvg_2.4.0 cellranger_1.1.0
## [75] rngtools_1.5.2 base64_2.0.1
## [77] Matrix_1.5-3 Rhdf5lib_1.20.0
## [79] base64enc_0.1-3 viridisLite_0.4.1
## [81] png_0.1-8 rjson_0.2.21
## [83] bitops_1.0-7 KernSmooth_2.23-20
## [85] rhdf5filters_1.10.0 blob_1.2.4
## [87] DelayedMatrixStats_1.20.0 doRNG_1.8.6
## [89] stringr_1.5.0 nor1mix_1.3-0
## [91] readr_2.1.4 jpeg_0.1-10
## [93] scales_1.2.1 memoise_2.0.1
## [95] magrittr_2.0.3 zlibbioc_1.44.0
## [97] compiler_4.2.2 BiocIO_1.8.0
## [99] illuminaio_0.40.0 Rsamtools_2.14.0
## [101] cli_3.6.0 DSS_2.46.0
## [103] htmlTable_2.4.1 Formula_1.2-5
## [105] MASS_7.3-58.3 tidyselect_1.2.0
## [107] stringi_1.7.12 highr_0.10
## [109] yaml_2.3.7 askpass_1.1
## [111] latticeExtra_0.6-30 sass_0.4.5
## [113] VariantAnnotation_1.44.0 tools_4.2.2
## [115] rstudioapi_0.14 foreign_0.8-84
## [117] bsseq_1.34.0 gridExtra_2.3
## [119] digest_0.6.31 BiocManager_1.30.20
## [121] shiny_1.7.4 quadprog_1.5-8
## [123] Rcpp_1.0.10 siggenes_1.72.0
## [125] BiocVersion_3.16.0 later_1.3.0
## [127] org.Hs.eg.db_3.16.0 httr_1.4.5
## [129] AnnotationDbi_1.60.0 biovizBase_1.46.0
## [131] colorspace_2.1-0 rvest_1.0.3
## [133] XML_3.99-0.13 splines_4.2.2
## [135] statmod_1.5.0 multtest_2.54.0
## [137] systemfonts_1.0.4 xtable_1.8-4
## [139] jsonlite_1.8.4 dynamicTreeCut_1.63-1
## [141] R6_2.5.1 echarts4r_0.4.4
## [143] Hmisc_5.0-1 pillar_1.8.1
## [145] htmltools_0.5.4 mime_0.12
## [147] glue_1.6.2 fastmap_1.1.1
## [149] BiocParallel_1.32.5 interactiveDisplayBase_1.36.0
## [151] beanplot_1.3.1 codetools_0.2-19
## [153] utf8_1.2.3 lattice_0.20-45
## [155] bslib_0.4.2 tibble_3.2.0
## [157] curl_5.0.0 gtools_3.9.4
## [159] openssl_2.0.6 interp_1.1-3
## [161] survival_3.5-5 rmarkdown_2.20
## [163] munsell_0.5.0 rhdf5_2.42.0
## [165] GenomeInfoDbData_1.2.9 HDF5Array_1.26.0
## [167] impute_1.72.3 gtable_0.3.2