You can use the "nodupkey" function. Ex: PROC SORT DATA = WORK.libname NODUPKEY; BY variable; RUN;   ... View more

@morganmetzger wrote: You can import CSV files by using an INFILE statement. Ex:  DATA Work.library; INFILE "your-path-to-file.csv" DELIMITER = ','; INPUT variables RUN;   Since CSV means comma separated variables you use ',' as your delimiter. But technically the delimiter can be any character(s) or space(s). The default behavior when you just change the DLM= option is that multiple delimiters are treated the same as one.  That is useful when the default delimiter is a space as usually people use extra spaces to make the fields line up vertically.  But in a CSV file that is not what you want.  In a CSV file two adjacent commas means the value for that field in the middle is empty.   Best to use the DSD option (which will imply that comma is the delimiter) because then both adjacent delimiters and quoted values are handled properly.  ... View more

For most things in SAS you would be better off having a separate variable such as Sample_number, and another variable that indicates Control and Treatment_a and assuming the body of the numbers are measuring the same property then one variable for the measurement.   Heatmaps use 2 variables for the dimensions. From your example I am not sure what makes most sense for the dimensions as each  segment needs one value from each dimension for location and optionally a value to assign a weight or frequency for setting the color of the blocks.   Here is a not terribly useful but simple example using a data set you should have available. proc sgplot data=sashelp.class; heatmap x=height y=weight / nxbins=4 nybins=3 showxbins showybins freq=age ; run; The above code uses the given data set with the student height as the xaxis, student weight as yaxis, nxbins and nybins set maximum number of bins to group the values of height and weight and the Showxbins and Showybins will have axis tick marks at the (may be approximate) center of the bin. the Freq=Age treats each record as between 11 and 16 records just to use the feature. Remove that bit and see the difference of appearance. There will be gradient on the right of the graph indicating the value the colors of the bins in heatmap indicate. ... View more