---标题：“第3天 - 基因表达”输出：html_document ---加载[geoquery]（https://biocidodder.org/package/geoquery）包。````{r}库（地理查询）```get the data``````````````````````````````````````````````````````````````````````````````gse20986 / gse20986_raw.tar“，exdir =”data“）```gunzipping数据```{r} cels = list.files（”数据/“，模式=”[gz]“）sapply（粘贴（”数据“，cels，sep =”/“），gunzip）```创建您的permodata``` {r} phenodata =矩阵（rep（list.files（”data“），2），ncol = 2）类（phenodata）Phenodata < - As.Data.frame（Phenodata）Colnames（Phenodata）< - C（“名称”，“文件名”）Penodata $ Targets < - C（“Iris”，“Retina”，“Retina”，“Iris”，“视网膜”，“虹膜”，“脉络膜”，“choroid”，“脉络膜”，“huvec”，“huvec”，“huvec”，“huvec”）write.table（phenodata，“data / phenodata.txt”，quote = f，sep =“\ t”，row.names = f）`````{r}库（simpleaffy）celfiles < - read.affy（covdesc =“penodata.txt”，path =“data”）boxplot（celfiles）图书馆（rcolorbrewer）cols = brewer.pal（8，“set1”）ESET < - exprs（Celfiles）样本< - celfiles $ targets colnames（ESET）colnames（ESET）< - Samples Boxplot（Celfille，Col = Cols，拉斯= 2）的距离< - DIST（吨（ESET），方法= “最大”）簇< - hclust（距离）情节（簇）``````{R}需要（simpleaffy）要求（affyPLM）celfiles.gcrma= GCRMA（CEFILES）PAR（mfrow = C（1,2））Boxplot（Celfiles.gcrma，Col = Col，LAS = 2，Main =“后标准化”）;boxplot（celfiles，col = cols，las = 2，main =“预归一化”）dev.off（）距离< - dist（t（exprs（celfiles.gcrma）），方法=“最大”）集群< - hclust（距离）绘图（群集）``````{R}库（林马）Phenodata样本< - as.factor（样本）设计< - model.matrix（〜0 +样本）colnames（设计）colnames（设计） <- c("choroid", "huvec", "iris", "retina") design contrast.matrix = makeContrasts( huvec_choroid = huvec - choroid, huvec_retina = huvec - retina, huvec_iris <- huvec - iris, levels = design) fit = lmFit(celfiles.gcrma, design) huvec_fit <- contrasts.fit(fit, contrast.matrix) huvec_ebay <- eBayes(huvec_fit) library(hgu133plus2.db) library(annotate) probenames.list <- rownames(topTable(huvec_ebay, number = 100000)) getsymbols <- getSYMBOL(probenames.list, "hgu133plus2") results <- topTable(huvec_ebay, number = 100000, coef = "huvec_choroid") results <- cbind(results, getsymbols) ``` To make thresholds ```{r} summary(results) results$threshold <- "1" a <- subset(results, adj.P.Val < 0.05 & logFC > 5) results[rownames(a), "threshold"] <- "2" b <- subset(results, adj.P.Val < 0.05 & logFC < -5) results[rownames(b), "threshold"] <- "3" table(results$threshold) ``` To make ggplot ```{r} library(ggplot2) volcano <- ggplot(data = results, aes(x = logFC, y = -1*log10(adj.P.Val), colour = threshold, label = getsymbols)) volcano <- volcano + geom_point() + scale_color_manual(values = c("black", "red", "green"), labels = c("Not Significant", "Upregulated", "Downregulated"), name = "Key/Legend") volcano + geom_text(data = subset(results, logFC > 5 & -1*log10(adj.P.Val) > 5), aes(x = logFC, y = -1*log10(adj.P.Val), colour = threshold, label = getsymbols) ) ``` [GEO Link](http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE20986)