options nofmterr nocenter ps=75 ls=250 formdlim='*';
libname in 'C:\Documents and Settings\pkraft\Desktop\Current PC Database';
%include 'C:\Documents and Settings\pkraft\Desktop\CoCo development\29jul05\BPC3_CaP_Utils.sas';

%let snps=ACTHR_001 ACTHR_002 ACTHR_003 ACTHR_004 ACTHR_005 ACTHR_006 
          CYP17_001 CYP17_002 CYP17_003 CYP17_011 CYP17_017 CYP17_019 CYP17_020 
          CYP17_023 CYP17_024 CYP19_001 CYP19_002 CYP19_003 CYP19_004 CYP19_005 CYP19_006 CYP19_007 
          CYP19_010 CYP19_011 CYP19_013 CYP19_014 CYP19_018 CYP19_023 CYP19_025 CYP19_027 CYP19_028 
          CYP19_031 CYP19_032 CYP19_033 CYP19_034 CYP19_036 CYP1B1_007 CYP1B1_018 CYP1B1_027 
          CYP1B1_031 CYP1B1_042 CYP1B1_059 CYP1B1_FLJ32954_01 CYP1B1_FLJ32954_02 DOCK5_001 ESR2_001 
          ESR2_003 ESR2_004 ESR2_006 ESR2_013 ESR2_014 GNRH1_101 GNRHR_101 GNRHR_102 GNRHR_103 
          GNRHR_104 GNRHR_105 GNRHR_106 GNRHR_107 HSD17B1_001 HSD17B1_005 HSD17B1_006 HSD17B1_009 
          IGF1_001 IGF1_002 IGF1_004 IGF1_006 IGF1_007 IGF1_010 IGF1_011 IGF1_014 IGF1_016 IGF1_019 
          IGF1_022 IGF1_027 IGF1_028 IGF1_030 IGFBP1_001 IGFBP1_003 IGFBP1_004 IGFBP1_009 
          IGFBP1_010 IGFBP1_011 IGFBP1_012 IGFBP1_015 IGFBP3_003 IGFBP3_004 IGFBP3_005 IGFBP3_007 
          IGFBP3_010 IGFBP3_011 IGFBP3_013 IGFBP3_014 PGR_004 PGR_009 PGR_010 PGR_011 PGR_012 
          PGR_013 PGR_016 PGR_017 PGR_018 PGR_019 PGR_020 PGR_024 PGR_025 PGR_026 PGR_028 PGR_029 
          PGR_030 PGR_031 POMC_001 POMC_002 POMC_003 POMC_004 POMC_005 POMC_006;

/* Create summary dataset - controls */

data temp;
   set in.cohort0306; 
   array snps {%numargs(&snps)} &snps;
   where caco=0;
   do i=1 to %numargs(&snps);
      if snps{i} eq '0 0' then snps{i}=' ';
      if snps{i} eq '9 9' then snps{i}=' ';
   end;
run;

proc sort data=temp;
   by subcoh;
run;

ods output MarkerSumm=a AlleleFreq=b;
proc allele data=temp genocol delimiter=' ';
   by subcoh;
   var &snps;
run;
ods output close;

proc sort data=b;
   by subcoh locus;
proc transpose data=b out=c prefix=freq;
   by subcoh locus;
   var freq;
data c;
   set c;
   drop _name_ _label_;
proc transpose data=b out=d prefix=allele;
   by subcoh locus;
   var allele;
data d;
   set d;
   drop _name_;
proc freq data=temp;
   tables subcoh / out=e;
 run;

proc sort data=a;
   by subcoh;
data AlleleSummary;
   merge a(keep=subcoh locus nindiv) e(drop=percent);
   by subcoh;
   completion=nindiv/COUNT;
   drop count;
proc sort data=AlleleSummary;
   by subcoh locus;
proc sort data=a;
   by subcoh locus;
data AlleleSummary;
   merge allelesummary a(keep=subcoh locus nallele) d c;
   by subcoh locus;
data AlleleSummary;
   merge allelesummary a(drop=nindiv nallele);
   by subcoh locus;
   delta=(het-2*freq1*freq2)/(2*freq1*freq2);
run;

proc datasets;
   delete a b c d e temp;
quit;

PROC EXPORT DATA= WORK.ALLELESUMMARY 
            OUTFILE= "C:\Documents and Settings\pkraft\Desktop\Current PC Database\AlleleSummary_Controls.csv" 
            DBMS=CSV REPLACE;
RUN;
quit;