ETLINK.OUT ANNOTATED OUTPUT FROM ETLINK USING TEST.DAT FOR INPUT ETLINK: Linkage disequilibrium analysis plant data source 1/26/90 No. isolates = 52 No. loci = 20 Total no. alleles = 72 FOR THE FOLLOWING DISTRIBUTION, LEWONTIN'S (1964, GENETICS 49:49-67) STANDARDIZED MEASURE OF LINKAGE DISEQUILIBRIUM, D', IS CALCULATED FOR EACH PAIR OF ALLELES AT TWO LOCI AND THE FREQUENCY IS TABULATED OVER ALL PAIRS OF LOCI. D' RANGES FROM -1.0 TO +1.0. FOR E.COLI POPULATIONS THE DISTRIBUTION OF D' IS TYPICALLY U-SHAPED BECAUSE MOST PAIRS OF ALLELES ARE IN COMPLETE ASSOCIATION (I.E. ONLY 3 OF THE POSSIBLE 4 HAPLO- TYPES FOR 2 ALLELES AT LOCI ARE OBSERVED). SEE HEDRICK AND THOMSON (1986, GENETICS 112:135-156) FOR MORE INFORMATION ABOUT THIS MEASURE. Frequency distribution of D prime -1.0 1216. .514 -.9 27. .011 -.8 14. .006 -.7 22. .009 -.6 15. .006 -.5 21. .009 -.4 38. .016 -.3 19. .008 -.2 23. .010 -.1 34. .014 .0 18. .008 .1 25. .011 .2 44. .019 .3 27. .011 .4 31. .013 .5 44. .019 .6 23. .010 .7 25. .011 .8 24. .010 .9 17. .007 1.0 661. .279 Total no. comparisons = 2368 ALL PAIRS OF ALLELES AT TWO LOCI THIS IS THE FREQUENCY DISTRIBUTION OF THE NUMBER OF MISMATCHES (I.E. NUMBER OF LOCI WITH DIFFERENT ALLELES) IN COMPARISON OF THE ELECTRO- MORPH PROFILES OF ALL PAIRS OF ISOLATES. K IS THE NUMBER OF MISMATCHES, N IS THE NUMBER OF ISOLATE-PAIRS OBSERVED THE DIFFER AT K LOCI, AND FREQ IS THE RELATIVE FREQUENCY. FOR M ISOLATES THERE ARE M*M PAIRWISE COMPARISONS. Mismatch distribution k n freq 0 178 .066 1 224 .083 2 86 .032 3 36 .013 4 112 .041 5 96 .036 6 224 .083 7 310 .115 8 460 .170 -- I.E. 460 PAIRS OF ISOLATES (17% OF ALL PAIRS) 9 352 .130 DIFFER AT 8 LOCI 10 346 .128 11 162 .060 12 72 .027 13 34 .013 14 12 .004 15 0 .000 16 0 .000 17 0 .000 18 0 .000 19 0 .000 20 0 .000 THE NEXT TABLE GIVES OBSERVED AND EXPECTED MOMENTS OF THE MISMATCH DISTRIBUTION AS SHOWN IN TABLE 1 OF WHITTAM ET AL. (1983, PROC. NATL. ACAD. SCI. 80:1751-1755). THE CALCULATIONS FOLLOW THE METHODS OF BROWN ET AL. (1980, GENETICS 96:523-536) WITH EXPECTED MOMENTS CALCULATED FROM EQUATIONS 3-5 (PAGE 525), OBSERVED MOMENTS OBTAINED BY METHOD B (PAGE 529) AND LOWER(L1) AND UPPER(L2) CONFIDENCE LIMITS FOR THE VARIANCE ASSUMING INDEPENDENCE OF LOCI TABULATED BY EQUATION 23 (PAGE 530) OBSERVED EXPECTED INDEX Moments M(i) u(i) X(i) Ia = X(2) i = 2 11.927 3.211 2.714 VARIANCE i = 3 -25.183 .330 -77.297 3RD CENTRAL MOMENT i = 4 346.973 30.353 10.431 4TH CENTRAL MOMENT L1 = 1.970 L2 = 4.453 SE Ia = .1933 95% CONFIDENCE LIMITS SE IaIS THE STANDARD ERROR ON THE INDEX AS PRESENTED BY MAYNARD SMTIH AND COWORKERS (1993, PROC. NATL. ACAD. SCI 90:4384) QSTAR.OUT ETLINK ALSO WRITES A FILE LISTING HEDRICK AND THOMSON Q* VALUES FOR EACH PAIR OF LOCI. Q* IS CALCULATED FROM HEDRICK AND THOMSON EQUATION 9 USING Q FROM EQUATION 8. Q IS DISTRIBUTED APPROXIMATELY AS A CHI SQUARE WITH DF DEGREES OF FREEDOM UNDER THE NULL HYPOTHESIS OF LINKAGE EQUILIBRIUM.