突变特征可视化--sigminer

背景介绍 

癌症基因组在其生命周期中由各种突变过程形成，这些过程源于外源性和细胞固有的DNA损伤，以及容易出错的DNA复制，产生了特征突变谱，称为突变特征。sigminer包，帮助用户从基因组改变记录中提取、分析和可视化签名，从而为癌症研究提供新的见解。

R包安装 BiocManager::install("sigminer")library(sigminer)     

结果解析 01   CopyNumber Object   首先用read_copynumber()读取数据，需要是具有以下信息的绝对拷贝数配置文件。

# 加载数据集load(system.file("extdata", "toy_segTab.RData", package = "sigminer", mustWork = TRUE))cn <- read_copynumber(segTabs, seg_cols = c("chromosome", "start", "end", "segVal"), genome_build = "hg19", complement = FALSE, verbose = TRUE) Profile show_cn_profile(cn, nrow = 2, ncol = 1)

show_cn_circos(cn, samples = 1)

show_cn_distribution(cn, mode = "ld")show_cn_distribution(cn, mode = "cd")

Signature Object  

操作signature

sig_extract() 或 sig_auto_extract() 的结果是一个带有 Signature 类的列表。 

library(maftools)tcgaAvailable()set.seed(1234) #brca <- readRDS("data/BRCA.RDs")brca <- tcga_load("BRCA")brca <- subsetMaf(brca, tsb = as.character(sample(brca@variants.per.sample$Tumor_Sample_Barcode, 100)))saveRDS(brca, file = "data/brca.rds")brca <- readRDS("data/brca.rds")mt_tally <- sig_tally( brca, ref_genome = "BSgenome.Hsapiens.UCSC.hg19", useSyn = TRUE)mt_sig <- sig_unify_extract(mt_tally$nmf_matrix, range = 10, nrun = 10)sig_signature(mt_sig2)[1:5, ]show_sig_profile(mt_sig, mode = "SBS", paint_axis_text = FALSE, x_label_angle = 90)

options(sigminer.copynumber.max = 20)# Load copy number objectload(system.file("extdata", "toy_copynumber.RData", package = "sigminer", mustWork = TRUE))# Use method designed by Wang, Shixiang et al.cn_tally_W <- sig_tally(cn, method = "W")sig_w <- sig_extract(cn_tally_W$nmf_matrix, n_sig = 2)show_sig_profile(sig_w, mode = "copynumber", normalize = "feature", method = "W", style = "cosmic")

show_sig_consensusmap(mt_sig)

简单分析流程   数据获取 library(sigminer)data("simulated_catalogs")mat <- t(simulated_catalogs$set1)mat[1:5, 1:5] 提取signature e1 <- bp_extract_signatures(mat, range = 8:12, n_bootstrap = 5, n_nmf_run = 10) 检查哪个signature号是正确的 bp_show_survey2(e1, highlight = 10)

获取10个signature obj <- bp_get_sig_obj(e1, 10) 可视化signature文件 show_sig_profile(obj, mode = "SBS", style = "cosmic")

show_sig_exposure(obj, rm_space = TRUE)

计算与 COSMIC 参考signature的相似度 sim <- get_sig_similarity(obj, sig_db = "SBS")if (require(pheatmap)) { pheatmap::pheatmap(sim$similarity)}

小编总结 作为最新发布的突变特征提取和可视化R包，sigminerd的使用是非常简单的，但是一定要注意输入数据的内容要包含关键信息。  

文字均为原创，欢迎读者分享或转发到朋友圈，任何公众号或其他媒体未经许可不得私自转载或抄袭。