meme suite —— Motif分析百宝箱（二）

时间：2024-01-03 18:10:41 买帖  | 投诉/举报

篇首语：本文由小编为大家整理，主要介绍了meme suite —— Motif分析百宝箱（二）相关的知识，希望对你有一定的参考价值。

参考技术A Motif Discovery中还包括MEME-ChIP，可对ChIP-seq或CLIP-seq数据的DNA序列进行一系列的Motif分析。该方法整合了：
1、MEME&STREME功能，用以预测Motif（de novo Motif）；
2、CentriMo用以寻找输入序列中间区段内已知的Motif，适合ChIP-Seq数据检出峰所在序列上富集的已知 Motif；
3、通过工具Tomtom功能比较已知的motif进行相似性分析，并对重要的motif进行分组；
4、Spamo与CentriMo功能相似，也是对Motif进行富集；
5、FIMO功能旨在输出Motif在基因的位置信息。

MEME-ChIP具体参数可以使用默认值，或按需求更改。示意图如下：

我们以网站Sample Output中MEME-ChIP example结果为例进行说明。

r 导入荷马ChIP-Seq Motif数据

# Daniel Cook 2014# Danielecook.com## Use this function to import ChIP Seq Data generated by Homer. This data is generated using homers findMotifsGenome.pl# command with the '-find <motif file>' argument. Generate output# looks like this:## 1. Peak/Region ID# 2. Chromosome# 3. Start# 4. End# 5. Strand of Peaks# 6-18: annotation information# 19. CpG%# 20. GC%# 21. Motif Instances <distance from center of region>(<sequence>,<strand>,<conservation>)## Getting the information from column 21 for plotting, for example, frequency can be tricky. This funciton aims to help that# by converting the data format from wide to long. Currently, annotation columns are ignored but you can modify the function to# fix.library(stringr)library(splitstackshape)library(reshape2)library(dplyr)suppressMessages(library(data.table))import_peak_file <- function(filename, peak_type) {  df <- fread(filename)  setnames(df, names(df)[1],"Peak")  setnames(df, names(df)[9:length(names(df))] , str_extract(names(df)[9:length(names(df))], "[A-Za-z0-9]+"))  motifs <- names(df)[9:length(names(df))]  # Reshape latter columns  # Remove extra columns if present  df$Annotation <- NULL  df$"Focus Ratio/Region Size" <- NULL  df$Detailed <- NULL  df$Distance <- NULL  df$Nearest <- NULL  df$Entrez <- NULL  df$Nearest <- NULL  df$Nearest <- NULL  df$Nearest <- NULL  df$Gene <- NULL  df$Gene <- NULL  df$Gene <- NULL  df$Gene <- NULL    # Melt Again  df <- melt(df, id.vars = 1:8)  df$value <- str_replace_all(df$value, "),", "|")  df <- filter(df, value != "")  df <- concat.split(as.data.frame(df), split.col = c("value"), sep="|")  df$value <- NULL    df <- rename(df, c("variable"="motif_name"))  df <- melt(df, id.vars = 1:9)  df$value_1 <- NULL  df$variable <- NULL  df <- filter(df, value != "")    df$MOTIF <- str_extract(df$value,"[ATCG]+")  df$POS <- as.integer(str_extract(df$value,"[-0-9]+"))  df$STRAND <- str_extract(df$value,"(,[+|-],)")  df$STRAND <- str_replace_all(df$STRAND,",","")  df$value <- NULL  df <- filter(df, !is.na(POS))  df$peak_type <- peak_type  df}

以上是关于meme suite —— Motif分析百宝箱（二）的主要内容，如果未能解决你的问题，请参考以下文章