Mercurial > repos > mingchen0919 > rmarkdown_deseq2_count_matrix

comparison rmarkdown_deseq2_count_matrix.Rmd @ 1:629323b5fc0c draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
update tool
author mingchen0919
date Sat, 30 Dec 2017 16:39:39 -0500
parents c1f718dd6c7a
children 8ceda5896765
comparison
equal deleted inserted replaced
0:c1f718dd6c7a 1:629323b5fc0c
14 error = TRUE 14 error = TRUE
15 ) 15 )
16 ``` 16 ```
17 17
18 18
19 ## User input 19 # User input
20 20
21 ```{r 'user input'} 21 ```{r 'user input'}
22 df = data.frame(name = names(opt)[-1], 22 df = data.frame(name = names(opt)[-1],
23 value = unlist(opt)) 23 value = unlist(opt))
24 df 24 datatable(df, rownames = FALSE)
25 ``` 25 ```
26 26
27 27
28 ## Count Matrix 28 # Count Matrix
29
30 Display the first 100 rows of count data matrix.
29 31
30 ```{r 'count matrix'} 32 ```{r 'count matrix'}
31 count_data = read.table(opt$count_data) 33 count_data = read.table(opt$count_data)
32 head(count_data, 10) 34 col_names = trimws(strsplit(opt$count_matrix_column_names, ',')[[1]])[1:ncol(count_data)]
35 col_names = col_names[!is.na(col_names)]
36 colnames(count_data)[1:length(col_names)] = col_names
37 datatable(head(count_data, 100))
33 ``` 38 ```
34 39
35 ```{r 'ste[ 2'} 40 # Column Data
36 41
42 ```{r 'column data'}
43 col_data = read.table(opt$col_data,
44 stringsAsFactors = FALSE, sep=',', header = TRUE, row.names = 1)
45 datatable(col_data)
46 ```
47
48 # Match sample names
49
50 The goal of this step is to rearrange the rows of the column data matrix so that the samples rows in the count data matrix and the sample columns in the count data matrix are in the same order.
51
52 ```{r 'match sample names'}
53 col_data = col_data[col_names, ]
54 datatable(col_data)
55 ```
56
57 # DESeqDataSet
58
59 ```{r 'DeseqDataSet'}
60 dds = DESeqDataSetFromMatrix(countData = count_data,
61 colData = col_data,
62 design = formula(opt$design_formula))
63 dds
64 ```
65
66 Pre-filter low count genes
67
68 ```{r 'pre-filtering'}
69 keep = rowSums(counts(dds)) >= 10
70 dds = dds[keep, ]
71 dds
72 ```
73
74 # Differential expression analysis
75
76 ```{r 'differential expression analysis'}
77 dds = DESeq(dds)
78 # res = results(dds, contrast = c(opt$contrast_condition, opt$treatment, opt$control))
79 res = results(dds)
80 resultsNames(dds)
81 if(nrow(res) > 500) {
82 cat('The result table has more than 500 rows. Only 500 rows are randomly selected to dispaly.')
83 datatable(as.data.frame(res)[sample(1:nrow(res), 500), ])
84 } else {
85 datatable(as.data.frame(res))
86 }
37 ``` 87 ```
38 88
39 89
90
91 ```{r 'write results into csv'}
92 #Write results into a CSV file.
93 write.csv(res, 'differential_genes.csv')
94 ```
95
96 # MAplot
97
98 ```{r}
99 plotMA(res)
100 ```
101
102
103 ```{r 'save R objects'}
104 # Save R objects to a file
105 save(dds, opt, file = 'deseq2.RData')
106 ```
107