nucleR-package(nucleR)
nucleR-package()所属R语言包：nucleR
Nucleosome positioning package for R
核小体定位为R包
译者：生物统计家园网 机器人LoveR
描述----------Description----------
Nucleosome positioning from Tiling Arrays and High-Troughput Sequencing Experiments
从平铺阵列和高Troughput的测序实验的核小体定位
Details
详情----------Details----------
This package provides a convenient pipeline to process and analize nucleosome positioning experiments from High-Troughtput Sequencing or Tiling Arrays.
这个软件包提供了一个方便的管道来处理和analize的高Troughtput测序或平铺阵列的核小体定位实验。
Despite it's use is intended to nucleosome experiments, it can be also useful for general ChIP experiments, such as ChIP-on-ChIP or ChIP-Seq.
尽管它的使用是为了核小体实验，它可以也很有用，如芯片上芯片或芯片的SEQ一般芯片实验。
See following example for a brief introduction to the available functions
看到下面的例子，简要介绍了可用的功能
作者（S）----------Author(s)----------
Oscar Flores
Maintainer: Oscar Flores <oflores@mmb.pcb.ub.es>
举例----------Examples----------
#Load example dataset:[负载例如数据集：]
# some NGS paired-end reads, mapped with Bowtie and processed with R[一些农工商配对末端读取与领结，映射和处理与R]
# it is a RangedData object with the start/end coordinates for each read.[它是为每个阅读的开始/结束的坐标RangedData对象。]
reads = get(data(nucleosome_htseq))
#Process the paired end reads, but discard those with length > 200[处理配对结束读取，但丢弃长度大于200]
preads_orig = processReads(reads, type="paired", fragmentLen=200)
#Process the reads, but now trim each read to 40bp around the dyad[处理的读取，但现在修剪每个读周围对子40BP]
preads_trim = processReads(reads, type="paired", fragmentLen=200, trim=40)
#Calculate the coverage, directly in reads per million (r.p.m)[计算的覆盖面，直接读取每百万（转）]
cover_orig = coverage.rpm(preads_orig)
cover_trim = coverage.rpm(preads_trim)
#Compare both coverages, the dyad is much more clear in trimmed version[比较这两种覆盖，更加清晰的对子是在修剪版本]
t1 = as.vector(cover_orig[[1]])[1:2000]
t2 = as.vector(cover_trim[[1]])[1:2000]
t1 = (t1-min(t1))/max(t1-min(t1)) #Normalization[标准化]
t2 = (t2-min(t2))/max(t2-min(t2)) #Normalization[标准化]
plot(t1, type="l", lwd="2", col="blue", main="Original vs Trimmed coverage")
lines(t2, lwd="2", col="red")
legend("bottomright", c("Original coverage", "Trimmed coverage"), lwd=2, col=c("blue","red"), bty="n")
#Let's try to call nucleosomes from the trimmed version[让我们尝试从修剪的版本调用的核小体]
#First of all, let's remove some noise with FFT[首先，让我们消除一些噪音与FFT]
#Power spectrum will be plotted, look how with a 2%[将绘制功率谱，2％]
#of the components we capture almost all the signal[的组成部分，我们几乎所有的信号捕获]
cover_clean = filterFFT(cover_trim, pcKeepComp=0.02, showPowerSpec=TRUE)
#How clean is now?[何洁现在是？]
plot(as.vector(cover_trim[[1]])[1:4000], t="l", lwd=2, col="red", main="Noisy vs Filtered coverage")
lines(cover_clean[[1]][1:4000], lwd=2, col="darkgreen")
legend("bottomright", c("Input coverage", "Filtered coverage"), lwd=2, col=c("red","darkgreen"), bty="n")
#And how similar? Let's see the correlation[多么相似？让我们来看看相关]
cor(cover_clean[[1]], as.vector(cover_trim[[1]]))
#Now it's time to call for peaks, first just as points[现在是呼吁峰时间，起初只是点]
#See that the score is only a measure of the height of the peak[看到比分是只有一峰高的措施]
peaks = peakDetection(cover_clean, threshold="25%", score=TRUE)
plotPeaks(peaks[[1]], cover_clean[[1]], threshold="25%")
#Do the same as previously, but now we will create the nucleosome calls:[做像以前一样，但现在我们将创建核小体的呼叫：]
peaks = peakDetection(cover_clean, width=147, threshold="25%", score=TRUE)
plotPeaks(peaks, cover_clean[[1]], threshold="25%")
#This is all. From here, you can filter, merge or work with the nucleosome[这就是一切。从这里，你可以过滤，合并或使用核小体]
#calls using standard IRanges functions and R/Bioconductor manipulation[要求使用标准IRanges功能和R / Bioconductor操纵]
转载请注明:出自 生物统计家园网(http://www.biostatistic.net)。
注：
注1：为了方便大家学习，本文档为生物统计家园网机器人LoveR翻译而成，仅供个人R语言学习参考使用，生物统计家园保留版权。
注2：由于是机器人自动翻译，难免有不准确之处，使用时仔细对照中、英文内容进行反复理解，可以帮助R语言的学习。
注3：如遇到不准确之处，请在本贴的后面进行回帖，我们会逐渐进行修订。
|
发表于 2012-2-26 07:54:50
