@@ -4,7 +4,7 @@ library(zoo)
44library(foreign )
55library(tiff )
66library(data.table )
7- source(' /users /jaywarrick/GoogleDrive/SingleCell/AndreaMatlabFiles/JayRFiles /preprocessingHelperFunctions.R'
7+ source(' /Users /jaywarrick/Public/Dropbox/GitHub/R-SingleCell /preprocessingHelperFunctions.R'
88
99# Generate Master data files based on arff files from JEX
1010# masterData <- getData()
@@ -24,91 +24,13 @@ rm(savedData)
2424# Remove ROIS with errors (bubbles etc.)
2525data <- data [! (data $ ID %in% errors $ ID ),]
2626
27- # Focus on M51R Data
28- data <- subset(data , Virus == ' M51R'
27+ # preprocess and write the preprocessed data to files
28+ temp <- preprocessVirusData(data , virusType = ' M51R'
29+ write.table(temp $ single , ' /Users/jaywarrick/Google Drive/SingleCellLatest/Processed Data/M51R_Data_1Cell.txt' row.names = FALSE )
30+ write.table(temp $ zero , ' /Users/jaywarrick/Google Drive/SingleCellLatest/Processed Data/M51R_Data_0Cell.txt' row.names = FALSE )
31+ write.table(temp $ thresholds , ' /Users/jaywarrick/Google Drive/SingleCellLatest/Processed Data/M51R_Thresholds.txt' row.names = FALSE )
32+ temp <- preprocessVirusData(data , virusType = ' N1'
33+ write.table(temp $ single , ' /Users/jaywarrick/Google Drive/SingleCellLatest/Processed Data/N1_Data_1Cell.txt' row.names = FALSE )
34+ write.table(temp $ zero , ' /Users/jaywarrick/Google Drive/SingleCellLatest/Processed Data/N1_Data_0Cell.txt' row.names = FALSE )
35+ write.table(temp $ thresholds , ' /Users/jaywarrick/Google Drive/SingleCellLatest/Processed Data/N1_Thresholds.txt' row.names = FALSE )
2936
30- # Read in the times at which each frame was acquired and fill in the time column with the image frame number (starting at 1)
31- timedata <- read.csv(' /Users/jaywarrick/Desktop/Octave/timeData.dat'
32- data [, time : = timedata $ Time [1 : length(Time )], by = ID ]
33-
34- # Calculate background corrected values for each color
35- data $ R.BC <- data $ R0_CellMax - data $ R0_Mode
36- data $ G.BC <- data $ G0_CellMax - data $ G0_Mode
37- data $ B.BC <- data $ B0_CellMax - data $ B0_Mode
38-
39- # Isolate data with 0 cell in the wells
40- null <- subset(data , Cell.Count == 0 )
41-
42- # Remove wells with more than 1 cell
43- data <- subset(data , ! is.na(Cell.Count ))
44-
45- # Get the mean null signals over time for each ROI and take the median for each
46- nullSummary <- null [, list (R.BC.Mean = mean(R.BC ),G.BC.Mean = mean(G.BC ),B.BC.Mean = mean(B.BC )), by = list (Device ,Image ,ID )] # Time-averaged mean of R, G, and B signals for each ROI with 0-cells in them
47- nullSummary <- nullSummary [, list (R.BC.Mean.Median = median(R.BC.Mean ), G.BC.Mean.Median = median(G.BC.Mean ), B.BC.Mean.Median = median(B.BC.Mean )), by = list (Device ,Image )] # Median of 0-cell well R, G, and B signals for each image and device
48-
49- # Subtract the median signal of null wells obtained for image from data in the corresponding images
50- nullSummary <- data.frame (nullSummary )
51- data [, R.NULL : = nullSummary [nullSummary $ Image == Image [1 ] & nullSummary $ Device == Device [1 ],]$ R.BC.Mean.Median , by = list (Device ,Image )]
52- data [, G.NULL : = nullSummary [nullSummary $ Image == Image [1 ] & nullSummary $ Device == Device [1 ],]$ G.BC.Mean.Median , by = list (Device ,Image )]
53- data [, B.NULL : = nullSummary [nullSummary $ Image == Image [1 ] & nullSummary $ Device == Device [1 ],]$ B.BC.Mean.Median , by = list (Device ,Image )]
54- data [, R.BC.NULL : = R.BC - R.NULL ]
55- data [, G.BC.NULL : = G.BC - G.NULL ]
56- data [, B.BC.NULL : = B.BC - B.NULL ]
57-
58- # Apply illumination correction
59- countData <- data.frame (countData )
60- data [, IF.Factor : = countData [countData $ ID == ID [1 ],]$ IF.Factor , by = ID ]
61- data [, R.Final : = R.BC.NULL / IF.Factor ]
62- data [, G.Final : = G.BC.NULL / IF.Factor ]
63- data [, B.Final : = B.BC.NULL / IF.Factor ]
64-
65- # Separate the data into data for 0-cell wells and 1-cell wells
66- single <- subset(data , Cell.Count == 1 )
67- zero <- subset(data , Cell.Count == 0 )
68-
69- # Create a quick plot for a sense check
70- plot(c(),c(),xlim = c(0 ,47 ),ylim = c(0 ,max(single $ R.Final )))
71- single [, lines(Time ,R.Final , col = ' red' by = ID ]
72- zero [, lines(Time ,R.Final , col = ' black' by = ID ]
73-
74- plot(c(),c(),xlim = c(0 ,47 ),ylim = c(0 ,max(single $ G.Final )))
75- single [, lines(Time ,G.Final , col = ' green' by = ID ]
76- zero [, lines(Time ,G.Final , col = ' black' by = ID ]
77-
78- plot(c(),c(),xlim = c(0 ,47 ),ylim = c(0 ,max(single $ B.Final )))
79- single [, lines(Time ,B.Final , col = ' blue' by = ID ]
80- zero [, lines(Time ,B.Final , col = ' black' by = ID ]
81-
82- # Determine the threshold based on 0-cell data
83- zeroSummary <- zero [, list (R.Mean = mean(R.Final ),G.Mean = mean(G.Final ),B.Mean = mean(B.Final )), by = list (Device ,Image ,ID )]
84- zeroSummary <- zeroSummary [, list (R.Mean.Median = median(R.Mean ), G.Mean.Median = median(G.Mean ), B.Mean.Median = median(B.Mean ), R.Mean.StdDev = sd(R.Mean ), G.Mean.StdDev = sd(G.Mean ), B.Mean.StdDev = sd(B.Mean )), by = list (Device ,Image )]
85- zeroSummary [, R.Thresh : = R.Mean.Median + 3 * R.Mean.StdDev ]
86- zeroSummary [, G.Thresh : = G.Mean.Median + 3 * G.Mean.StdDev ]
87- zeroSummary [, B.Thresh : = B.Mean.Median + 3 * B.Mean.StdDev ]
88- thresholds <- zeroSummary [, list (R = max(R.Thresh ), G = max(G.Thresh ), B = max(B.Thresh ))]
89-
90- # Zero points below threshold and spurrious points above threshold (i.e. less than 4 points in a row above 0)
91- single [,R : = rollmin(R.Final , thresh = thresholds $ R ),by = ID ]
92- single [,G : = rollmin(G.Final , thresh = thresholds $ G ),by = ID ]
93- single [,B : = rollmin(B.Final , thresh = thresholds $ B ),by = ID ]
94-
95- # For each color, mark trajectories that have a signal
96- hasSignal <- function (piece ){if (max(piece )> 0 ){return (TRUE )}else {return (FALSE )}}
97- single [,Flag.R : = hasSignal(R ),by = ID ]
98- single [,Flag.G : = hasSignal(G ),by = ID ]
99- single [,Flag.B : = hasSignal(B ),by = ID ]
100-
101- # Create a quick plot for a sense check
102- plot(c(),c(),xlim = c(0 ,47 ),ylim = c(1 ,max(single $ R )), log = ' y'
103- single [, lines(Time ,R , col = rgb(1 ,0 ,0 ,0.2 )), by = ID ]
104-
105- plot(c(),c(),xlim = c(0 ,47 ),ylim = c(1 ,max(single $ G )), log = ' y'
106- single [, lines(Time ,G , col = rgb(0 ,1 ,0 ,0.2 )), by = ID ]
107-
108- plot(c(),c(),xlim = c(0 ,47 ),ylim = c(1 ,max(single $ B )), log = ' y'
109- single [, lines(Time ,B , col = rgb(0 ,0 ,1 ,0.2 )), by = ID ]
110-
111- # Write the preprocessed data to a file
112- write.table(single , ' /Users/jaywarrick/GoogleDrive/SingleCell/AndreaMatlabFiles/JayRFiles/M51R_Data_1Cell.txt' row.names = FALSE )
113- write.table(zero , ' /Users/jaywarrick/GoogleDrive/SingleCell/AndreaMatlabFiles/JayRFiles/M51R_Data_0Cell.txt' row.names = FALSE )
114- write.table(thresholds , ' /Users/jaywarrick/GoogleDrive/SingleCell/AndreaMatlabFiles/JayRFiles/M51R_Thresholds.txt' row.names = FALSE )
0 commit comments