In [40]:

    

library(ggplot2)


    

In [22]:

    

# Get the data
local_path = '/home/irockafe/Dropbox (MIT)/Alm_Lab/projects/'
project_path = paste('/revo_healthcare/data/processed/Husermet_MTBLS97/',
                    'positive_ion/',sep='')
total_path = paste(local_path, project_path,sep='')
# load feature table
# patients x features
feats = read.csv(paste(total_path,
               'Husermet_UPLCMS_positive_ion_mode_featuretable.csv',
                   sep=''))
metadata = read.csv(paste(total_path,
                    'Husermet_UPLCMS_positive_ion_mode_metadata.csv',
                    sep=''))


    

In [94]:

    

head(meta)
dim(metadata)
columns = colnames(metadata)
hist_columns = c(columns[2], tail(columns,28))
hist_columns


    

    






Idx
Age
Gender
BMI
SBP
DBP
CSmoker
ALB
PROT
CREA
⋯
PLT
RBC
WBC
TBIL
ALP
ALT
AST
GGT
PHOS
LDH

	
1       
60.00000
M       
25.01000
132     
78      
N       
NA      
72.0    
 93     
⋯       
 NA     
NA      
NA      
10.0    
 66.0   
49      
41      
23.0    
0.97    
465     
	
2       
54.33333
M       
24.38000
140     
80      
N       
NA      
66.0    
 70     
⋯       
 NA     
NA      
NA      
24.0    
 70.0   
25      
19      
36.0    
1.02    
332     
	
3       
56.75000
M       
28.87000
142     
76      
E       
NA      
70.0    
106     
⋯       
 NA     
NA      
NA      
 9.0    
 62.0   
18      
19      
20.0    
1.00    
322     
	
4       
41.20000
M       
24.10000
135     
79      
U       
42      
70.9    
 89     
⋯       
247     
 5      
 8      
 4.1    
 37.1   
18      
14      
16.3    
  NA    
 NA     
	
5       
55.75000
M       
28.90842
160     
98      
N       
44      
  NA    
113     
⋯       
203     
 5      
 5      
  NA    
   NA   
NA      
NA      
  NA    
  NA    
 NA     
	
6       
35.08333
M       
27.28000
110     
60      
E       
NA      
69.0    
 77     
⋯       
 NA     
NA      
NA      
 7.0    
161.0   
40      
24      
34.0    
0.92    
318     









    






	
1189
	
31








    






	
'Age'
	
'BMI'
	
'SBP'
	
'DBP'
	
'CSmoker'
	
'ALB'
	
'PROT'
	
'CREA'
	
'SGLUC'
	
'SODIUM'
	
'K'
	
'CALCIUM'
	
'CHOL'
	
'TRIG'
	
'HDLC'
	
'LDLC'
	
'Ratio'
	
'UREA'
	
'HAEM'
	
'PLT'
	
'RBC'
	
'WBC'
	
'TBIL'
	
'ALP'
	
'ALT'
	
'AST'
	
'GGT'
	
'PHOS'
	
'LDH'

In [96]:

    

# Show me an age breakdown
# TODO make a gridplot of all the distributions
for (i in hist_columns){
    print(i)
    hist(metadata[[i]], xlab=i,  main=sprintf("Histogram of %s", i))
    }


    

    




[1] "Age"
[1] "BMI"






    













    




[1] "SBP"






    













    




[1] "DBP"
[1] "CSmoker"






    




Error in hist.default(metadata[[i]], xlab = i, main = sprintf("Histogram of %s", : 'x' must be numeric
Traceback:

1. hist(metadata[[i]], xlab = i, main = sprintf("Histogram of %s", 
 .     i))   # at line 5 of file <text>
2. hist.default(metadata[[i]], xlab = i, main = sprintf("Histogram of %s", 
 .     i))
3. stop("'x' must be numeric")





    













    

In [48]:

    

p <- ggplot(mpg, aes(displ, cty)) + geom_point()
p