Customizing oncoplots
Anand Mayakonda
2024-05-01
#path to TCGA LAML MAF file
laml.maf = system.file('extdata', 'tcga_laml.maf.gz', package = 'maftools')
#clinical information containing survival information and histology. This is optional
laml.clin = system.file('extdata', 'tcga_laml_annot.tsv', package = 'maftools')
laml = read.maf(maf = laml.maf,
clinicalData = laml.clin,
verbose = FALSE)0.1 Including Transition/Transversions into oncoplot
0.2 Changing colors for variant classifications
#One can use any colors, here in this example color palette from RColorBrewer package is used
vc_cols = RColorBrewer::brewer.pal(n = 8, name = 'Paired')
names(vc_cols) = c(
'Frame_Shift_Del',
'Missense_Mutation',
'Nonsense_Mutation',
'Multi_Hit',
'Frame_Shift_Ins',
'In_Frame_Ins',
'Splice_Site',
'In_Frame_Del'
)
print(vc_cols)
#> Frame_Shift_Del Missense_Mutation Nonsense_Mutation Multi_Hit
#> "#A6CEE3" "#1F78B4" "#B2DF8A" "#33A02C"
#> Frame_Shift_Ins In_Frame_Ins Splice_Site In_Frame_Del
#> "#FB9A99" "#E31A1C" "#FDBF6F" "#FF7F00"
oncoplot(maf = laml, colors = vc_cols, top = 10)
0.3 Including copy number data into oncoplots.
There are two ways one include CN status into MAF. 1. GISTIC results 2. Custom copy number table
0.3.1 GISTIC results
Most widely used tool for copy number analysis from large scale studies is GISTIC and we can simultaneously read gistic results along with MAF. GISTIC generates numerous files but we need mainly four files all_lesions.conf_XX.txt, amp_genes.conf_XX.txt, del_genes.conf_XX.txt, scores.gistic where XX is confidence level. These files contain significantly altered genomic regions along with amplified and deleted genes respectively.
#GISTIC results LAML
all.lesions =
system.file("extdata", "all_lesions.conf_99.txt", package = "maftools")
amp.genes =
system.file("extdata", "amp_genes.conf_99.txt", package = "maftools")
del.genes =
system.file("extdata", "del_genes.conf_99.txt", package = "maftools")
scores.gis =
system.file("extdata", "scores.gistic", package = "maftools")
#Read GISTIC results along with MAF
laml.plus.gistic = read.maf(
maf = laml.maf,
gisticAllLesionsFile = all.lesions,
gisticAmpGenesFile = amp.genes,
gisticDelGenesFile = del.genes,
gisticScoresFile = scores.gis,
isTCGA = TRUE,
verbose = FALSE,
clinicalData = laml.clin
)
This plot shows frequent deletions in TP53 gene which is located on one of the significantly deleted locus 17p13.2.
0.3.2 Custom copy-number table
In case there is no GISTIC results available, one can generate a table containing CN status for known genes in known samples. This can be easily created and read along with MAF file.
For example lets create a dummy CN alterations for DNMT3A in random 20 samples.
set.seed(seed = 1024)
barcodes = as.character(getSampleSummary(x = laml)[,Tumor_Sample_Barcode])
#Random 20 samples
dummy.samples = sample(x = barcodes,
size = 20,
replace = FALSE)
#Genarate random CN status for above samples
cn.status = sample(
x = c('ShallowAmp', 'DeepDel', 'Del', 'Amp'),
size = length(dummy.samples),
replace = TRUE
)
custom.cn.data = data.frame(
Gene = "DNMT3A",
Sample_name = dummy.samples,
CN = cn.status,
stringsAsFactors = FALSE
)
head(custom.cn.data)
#> Gene Sample_name CN
#> 1 DNMT3A TCGA-AB-2898 ShallowAmp
#> 2 DNMT3A TCGA-AB-2879 Del
#> 3 DNMT3A TCGA-AB-2920 Amp
#> 4 DNMT3A TCGA-AB-2866 Del
#> 5 DNMT3A TCGA-AB-2892 Del
#> 6 DNMT3A TCGA-AB-2863 ShallowAmp
laml.plus.cn = read.maf(maf = laml.maf,
cnTable = custom.cn.data,
verbose = FALSE)
oncoplot(maf = laml.plus.cn, top = 5)
0.4 Bar plots
leftBarData, rightBarData and topBarData arguments can be used to display additional values as barplots. Below example demonstrates adding gene expression values and mutsig q-values as left and right side bars respectivelly.
#Selected AML driver genes
aml_genes = c("TP53", "WT1", "PHF6", "DNMT3A", "DNMT3B", "TET1", "TET2", "IDH1", "IDH2", "FLT3", "KIT", "KRAS", "NRAS", "RUNX1", "CEBPA", "ASXL1", "EZH2", "KDM6A")
#Variant allele frequcnies (Right bar plot)
aml_genes_vaf = subsetMaf(maf = laml, genes = aml_genes, fields = "i_TumorVAF_WU", mafObj = FALSE)[,mean(i_TumorVAF_WU, na.rm = TRUE), Hugo_Symbol]
colnames(aml_genes_vaf)[2] = "VAF"
head(aml_genes_vaf)
#> Hugo_Symbol VAF
#> <char> <num>
#> 1: ASXL1 37.11250
#> 2: CEBPA 22.00235
#> 3: DNMT3A 43.51556
#> 4: DNMT3B 37.14000
#> 5: EZH2 68.88500
#> 6: FLT3 34.60294
#MutSig results (Right bar plot)
laml.mutsig = system.file("extdata", "LAML_sig_genes.txt.gz", package = "maftools")
laml.mutsig = data.table::fread(input = laml.mutsig)[,.(gene, q)]
laml.mutsig[,q := -log10(q)] #transoform to log10
head(laml.mutsig)
#> gene q
#> <char> <num>
#> 1: FLT3 12.64176
#> 2: DNMT3A 12.64176
#> 3: NPM1 12.64176
#> 4: IDH2 12.64176
#> 5: IDH1 12.64176
#> 6: TET2 12.64176
oncoplot(
maf = laml,
genes = aml_genes,
leftBarData = aml_genes_vaf,
leftBarLims = c(0, 100),
rightBarData = laml.mutsig,
rightBarLims = c(0, 20)
)
0.5 Including annotations
Annotations are stored in clinical.data slot of MAF.
getClinicalData(x = laml)
#> Tumor_Sample_Barcode FAB_classification days_to_last_followup
#> <char> <char> <char>
#> 1: TCGA-AB-2802 M4 365
#> 2: TCGA-AB-2803 M3 792
#> 3: TCGA-AB-2804 M3 2557
#> 4: TCGA-AB-2805 M0 577
#> 5: TCGA-AB-2806 M1 945
#> ---
#> 189: TCGA-AB-3007 M3 1581
#> 190: TCGA-AB-3008 M1 822
#> 191: TCGA-AB-3009 M4 577
#> 192: TCGA-AB-3011 M1 1885
#> 193: TCGA-AB-3012 M3 1887
#> Overall_Survival_Status
#> <char>
#> 1: 1
#> 2: 1
#> 3: 0
#> 4: 1
#> 5: 1
#> ---
#> 189: 0
#> 190: 1
#> 191: 1
#> 192: 0
#> 193: 0Include FAB_classification from clinical data as one of the sample annotations.
More than one annotations can be included by passing them to the argument clinicalFeatures. Above plot can be further enhanced by sorting according to annotations. Custom colors can be specified as a list of named vectors for each levels.
#Color coding for FAB classification
fabcolors = RColorBrewer::brewer.pal(n = 8,name = 'Spectral')
names(fabcolors) = c("M0", "M1", "M2", "M3", "M4", "M5", "M6", "M7")
fabcolors = list(FAB_classification = fabcolors)
print(fabcolors)
#> $FAB_classification
#> M0 M1 M2 M3 M4 M5 M6 M7
#> "#D53E4F" "#F46D43" "#FDAE61" "#FEE08B" "#E6F598" "#ABDDA4" "#66C2A5" "#3288BD"
oncoplot(
maf = laml, genes = aml_genes,
clinicalFeatures = 'FAB_classification',
sortByAnnotation = TRUE,
annotationColor = fabcolors
)
0.6 Highlighting samples
If you prefer to highlight mutations by a specific attribute, you can use additionalFeature argument.
Example: Highlight all mutations where alt allele is C.
Note that first argument (Tumor_Seq_Allele2) must a be column in MAF file, and second argument (C) is a value in that column. If you want to know what columns are present in the MAF file, use getFields.
getFields(x = laml)
#> [1] "Hugo_Symbol" "Entrez_Gene_Id" "Center"
#> [4] "NCBI_Build" "Chromosome" "Start_Position"
#> [7] "End_Position" "Strand" "Variant_Classification"
#> [10] "Variant_Type" "Reference_Allele" "Tumor_Seq_Allele1"
#> [13] "Tumor_Seq_Allele2" "Tumor_Sample_Barcode" "Protein_Change"
#> [16] "i_TumorVAF_WU" "i_transcript_name"0.7 Group by Pathways
Genes can be auto grouped based on their pathway belongings. Currently maftools has two pathway databases,
- oncogenic signalling pathways: A curated list of 10 signalling pathways
- Catalog of known driver genes classified by their biological processes
By setting pathways argument either sigpw or smgbp - cohort can be summarized by altered pathways. pathways argument also accepts a custom pathway list in the form of a two column tsv file or a data.frame containing gene names and their corresponding pathway.
0.7.1 Oncogenic siganlling pathways
setting pathways = 'sigpw' to draw 5 most affected pathways
oncoplot(maf = laml, pathways = "sigpw", gene_mar = 8, fontSize = 0.6, topPathways = 5)
#> Summarizing signalling pathways [Sanchez-Vega et al., https://doi.org/10.1016/j.cell.2018.03.035]
#> Drawing upto top 5 mutated pathways
#> Pathway N n_affected_genes fraction_affected Mutated_samples
#> <char> <int> <int> <num> <num>
#> 1: RTK-RAS 85 18 0.21176471 97
#> 2: TP53 6 2 0.33333333 15
#> 3: NOTCH 71 6 0.08450704 8
#> 4: Hippo 38 7 0.18421053 7
#> 5: WNT 68 3 0.04411765 4
#> 6: MYC 13 3 0.23076923 3
#> 7: PI3K 29 1 0.03448276 1
#> 8: NRF2 3 1 0.33333333 1
#> 9: Cell_Cycle 15 0 0.00000000 0
#> 10: TGF-Beta 7 0 0.00000000 0
#> Fraction_mutated_samples
#> <num>
#> 1: 0.502590674
#> 2: 0.077720207
#> 3: 0.041450777
#> 4: 0.036269430
#> 5: 0.020725389
#> 6: 0.015544041
#> 7: 0.005181347
#> 8: 0.005181347
#> 9: 0.000000000
#> 10: 0.000000000
0.7.2 Biological processes of known drivers
oncoplot(maf = laml, pathways = "smgbp", gene_mar = 8, fontSize = 0.8, topPathways = 5)
#> Summarizing known driver genes [Bailey et al., https://doi.org/10.1016/j.cell.2018.02.060]
#> Drawing upto top 5 mutated pathways
#> Pathway N n_affected_genes fraction_affected
#> <char> <int> <int> <num>
#> 1: RTK signaling 16 6 0.37500000
#> 2: Epigenetics DNA modifiers 1 1 1.00000000
#> 3: Chromatin other 14 7 0.50000000
#> 4: Metabolism 2 2 1.00000000
#> 5: Transcription factor 39 11 0.28205128
#> 6: Genome integrity 14 5 0.35714286
#> 7: MAPK signaling 9 3 0.33333333
#> 8: Other signaling 28 5 0.17857143
#> 9: Splicing 6 3 0.50000000
#> 10: Histone modification 3 2 0.66666667
#> 11: Other 22 3 0.13636364
#> 12: Protein homeostasis/ubiquitination 15 3 0.20000000
#> 13: Immune signaling 10 3 0.30000000
#> 14: Chromatin SWI/SNF complex 8 2 0.25000000
#> 15: RNA abundance 15 1 0.06666667
#> 16: PI3K signaling 9 1 0.11111111
#> 17: NOTCH signaling 1 1 1.00000000
#> 18: Apoptosis 3 0 0.00000000
#> 19: Cell cycle 8 0 0.00000000
#> 20: Chromatin histone modifiers 15 0 0.00000000
#> 21: NFKB signaling 2 0 0.00000000
#> 22: TGFB signaling 7 0 0.00000000
#> 23: TOR signaling 3 0 0.00000000
#> 24: Wnt/B-catenin signaling 8 0 0.00000000
#> Pathway N n_affected_genes fraction_affected
#> Mutated_samples Fraction_mutated_samples
#> <num> <num>
#> 1: 63 0.326424870
#> 2: 48 0.248704663
#> 3: 45 0.233160622
#> 4: 37 0.191709845
#> 5: 35 0.181347150
#> 6: 28 0.145077720
#> 7: 25 0.129533679
#> 8: 13 0.067357513
#> 9: 11 0.056994819
#> 10: 4 0.020725389
#> 11: 4 0.020725389
#> 12: 4 0.020725389
#> 13: 3 0.015544041
#> 14: 2 0.010362694
#> 15: 1 0.005181347
#> 16: 1 0.005181347
#> 17: 1 0.005181347
#> 18: 0 0.000000000
#> 19: 0 0.000000000
#> 20: 0 0.000000000
#> 21: 0 0.000000000
#> 22: 0 0.000000000
#> 23: 0 0.000000000
#> 24: 0 0.000000000
#> Mutated_samples Fraction_mutated_samples
0.7.3 Custom pathway list
pathways = data.frame(
Genes = c(
"TP53",
"WT1",
"PHF6",
"DNMT3A",
"DNMT3B",
"TET1",
"TET2",
"IDH1",
"IDH2",
"FLT3",
"KIT",
"KRAS",
"NRAS",
"RUNX1",
"CEBPA",
"ASXL1",
"EZH2",
"KDM6A"
),
Pathway = rep(c(
"TSG", "DNAm", "Signalling", "TFs", "ChromMod"
), c(3, 6, 4, 2, 3)),
stringsAsFactors = FALSE
)
head(pathways)
#> Genes Pathway
#> 1 TP53 TSG
#> 2 WT1 TSG
#> 3 PHF6 TSG
#> 4 DNMT3A DNAm
#> 5 DNMT3B DNAm
#> 6 TET1 DNAm
oncoplot(maf = laml, pathways = pathways, gene_mar = 8, fontSize = 0.6)
0.7.4 Collapse pathways
By setting collapsePathway = TRUE..
oncoplot(maf = laml, pathways = "sigpw", gene_mar = 8, fontSize = 0.6, topPathways = 5, collapsePathway = TRUE)
#> Summarizing signalling pathways [Sanchez-Vega et al., https://doi.org/10.1016/j.cell.2018.03.035]
#> Drawing upto top 5 mutated pathways
#> Pathway N n_affected_genes fraction_affected Mutated_samples
#> <char> <int> <int> <num> <num>
#> 1: RTK-RAS 85 18 0.21176471 97
#> 2: TP53 6 2 0.33333333 15
#> 3: NOTCH 71 6 0.08450704 8
#> 4: Hippo 38 7 0.18421053 7
#> 5: WNT 68 3 0.04411765 4
#> 6: MYC 13 3 0.23076923 3
#> 7: PI3K 29 1 0.03448276 1
#> 8: NRF2 3 1 0.33333333 1
#> 9: Cell_Cycle 15 0 0.00000000 0
#> 10: TGF-Beta 7 0 0.00000000 0
#> Fraction_mutated_samples
#> <num>
#> 1: 0.502590674
#> 2: 0.077720207
#> 3: 0.041450777
#> 4: 0.036269430
#> 5: 0.020725389
#> 6: 0.015544041
#> 7: 0.005181347
#> 8: 0.005181347
#> 9: 0.000000000
#> 10: 0.000000000
0.8 Combining everything
oncoplot(
maf = laml.plus.gistic,
draw_titv = TRUE,
pathways = pathways,
clinicalFeatures = c('FAB_classification', 'Overall_Survival_Status'),
sortByAnnotation = TRUE,
additionalFeature = c("Tumor_Seq_Allele2", "C"),
leftBarData = aml_genes_vaf,
leftBarLims = c(0, 100),
rightBarData = laml.mutsig[,.(gene, q)],
)
0.9 SessionInfo
sessionInfo()
#> R version 4.4.0 beta (2024-04-15 r86425)
#> Platform: x86_64-pc-linux-gnu
#> Running under: Ubuntu 22.04.4 LTS
#>
#> Matrix products: default
#> BLAS: /home/biocbuild/bbs-3.19-bioc/R/lib/libRblas.so
#> 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
#>
#> time zone: America/New_York
#> tzcode source: system (glibc)
#>
#> attached base packages:
#> [1] parallel stats4 stats graphics grDevices utils datasets
#> [8] methods base
#>
#> other attached packages:
#> [1] pheatmap_1.0.12 doParallel_1.0.17
#> [3] iterators_1.0.14 foreach_1.5.2
#> [5] NMF_0.27 bigmemory_4.6.4
#> [7] Biobase_2.64.0 cluster_2.1.6
#> [9] rngtools_1.5.2 registry_0.5-1
#> [11] BSgenome.Hsapiens.UCSC.hg19_1.4.3 BSgenome_1.72.0
#> [13] rtracklayer_1.64.0 BiocIO_1.14.0
#> [15] Biostrings_2.72.0 XVector_0.44.0
#> [17] GenomicRanges_1.56.0 GenomeInfoDb_1.40.0
#> [19] IRanges_2.38.0 S4Vectors_0.42.0
#> [21] BiocGenerics_0.50.0 mclust_6.1.1
#> [23] maftools_2.20.0
#>
#> loaded via a namespace (and not attached):
#> [1] tidyselect_1.2.1 gridBase_0.4-7
#> [3] dplyr_1.1.4 R.utils_2.12.3
#> [5] bitops_1.0-7 RaggedExperiment_1.28.0
#> [7] fastmap_1.1.1 RCurl_1.98-1.14
#> [9] GenomicAlignments_1.40.0 XML_3.99-0.16.1
#> [11] digest_0.6.35 lifecycle_1.0.4
#> [13] survival_3.6-4 magrittr_2.0.3
#> [15] compiler_4.4.0 rlang_1.1.3
#> [17] sass_0.4.9 tools_4.4.0
#> [19] utf8_1.2.4 yaml_2.3.8
#> [21] data.table_1.15.4 knitr_1.46
#> [23] S4Arrays_1.4.0 curl_5.2.1
#> [25] DelayedArray_0.30.0 plyr_1.8.9
#> [27] RColorBrewer_1.1-3 abind_1.4-5
#> [29] BiocParallel_1.38.0 R.oo_1.26.0
#> [31] grid_4.4.0 fansi_1.0.6
#> [33] colorspace_2.1-0 ggplot2_3.5.1
#> [35] scales_1.3.0 MultiAssayExperiment_1.30.0
#> [37] SummarizedExperiment_1.34.0 cli_3.6.2
#> [39] rmarkdown_2.26 crayon_1.5.2
#> [41] generics_0.1.3 bigmemory.sri_0.1.8
#> [43] reshape2_1.4.4 httr_1.4.7
#> [45] rjson_0.2.21 DNAcopy_1.78.0
#> [47] cachem_1.0.8 stringr_1.5.1
#> [49] zlibbioc_1.50.0 splines_4.4.0
#> [51] restfulr_0.0.15 matrixStats_1.3.0
#> [53] vctrs_0.6.5 Matrix_1.7-0
#> [55] jsonlite_1.8.8 berryFunctions_1.22.5
#> [57] jquerylib_0.1.4 glue_1.7.0
#> [59] codetools_0.2-20 stringi_1.8.3
#> [61] gtable_0.3.5 UCSC.utils_1.0.0
#> [63] munsell_0.5.1 tibble_3.2.1
#> [65] pillar_1.9.0 htmltools_0.5.8.1
#> [67] GenomeInfoDbData_1.2.12 R6_2.5.1
#> [69] evaluate_0.23 lattice_0.22-6
#> [71] highr_0.10 R.methodsS3_1.8.2
#> [73] Rsamtools_2.20.0 bslib_0.7.0
#> [75] uuid_1.2-0 Rcpp_1.0.12
#> [77] SparseArray_1.4.0 xfun_0.43
#> [79] MatrixGenerics_1.16.0 pkgconfig_2.0.3