! ! QTL via weighted likelihood of three covariance matrices ! The following MX script lines were read for group 1 #DEFINE NVAR 1 #DEFINE STARTQPAR 3 ! NVAR*(NVAR+1) +1 #DEFINE ENDQPAR 3 ! NVAR*(NVAR+2) #DEFINE LOCATION 1850 #NGROUPS 2 Note: #NGroup set number of groups to 2 TITLE QTL MODEL WITH HETEROGENEITY AND WEIGHTS DATA NINPUT=7 NMODEL=3 RECTANGULAR FILE=MXGH2.REC Rectangular continuous data read initiated NOTE: Rectangular file contained 600 records with data that contained a total of 4200 observations LABELS FAMID SIB1 SIB2 P2 P1 P0 PIHAT SELECT SIB1 SIB2 P0 P1 P2 ; DEFINITION P0 P1 P2 ; NOTE: Selection yields 600 data vectors for analysis NOTE: Vectors contain a total of 3000 observations NOTE: Definition yields 600 data vectors for analysis NOTE: Vectors contain a total of 1200 observations MATRICES A LOWER NVAR NVAR FREE ! BACKGROUND A (OR C) E LOWER NVAR NVAR FREE ! RANDOM ENVT K FULL 3 1 ! P(I)'S GO HERE Q FULL NVAR 1 FREE ! QTL EFFECT H FULL 1 1 ! .5 M FULL 1 NVAR FREE ! FOR MEAN I UNIT 3 2 ! TO SPREAD MEAN OUT FOR ALL MODELS & SIBS END MATRICES MATRIX H .5 SPECIFY K P0 P1 P2 ; MATRIX A .7 MATRIX E .7 MATRIX Q .7 START 0 M 1 1 TO M 1 NVAR BEGIN ALGEBRA; V = A*A' + Q*Q' + E*E' ; ! VARIANCE U = H@A*A' ; ! IBD 0 COV (=NON-QTL GENETIC COV) X = U + H@Q*Q' ; ! IBD 1 COV W = U + Q*Q' ; ! IBD 2 COV Z = V|U_U|V; ! IBD 0 MATRIX O = V|X_X|V; ! IBD 1 MATRIX T = V|W_W|V; ! IBD 2 MATRIX Y = Z_O_T; N = M|M; R = A|Q|E; END ALGEBRA; MEANS I@M ; COVARIANCE Y ; WEIGHTS K ; BOUND .01 100 E 1 1 OPTION RS SAMECOV IT=250 OPTION MU ISSAT ! option samecov tells it not to re-invert cov matrix as it doesn't change OPTION MXR=QTLEXP.MAT OPTION APPEND OPTION FORMAT=(7(F13.6,1X)) END The following MX script lines were read for group 2 GROUP 2: MODEL PIHAT DEVIATIONS AS A FUNCTION OF PHENOTYPES DATA NINPUT=7 RECTANGULAR FILE=MXGH2.REC Rectangular continuous data read initiated NOTE: Rectangular file contained 600 records with data that contained a total of 4200 observations LABELS FAMID SIB1 SIB2 P2 P1 P0 PIHAT SELECT SIB1 SIB2 PIHAT ; DEFINITION SIB1 SIB2 ; NOTE: Selection yields 600 data vectors for analysis NOTE: Vectors contain a total of 1800 observations NOTE: Definition yields 600 data vectors for analysis NOTE: Vectors contain a total of 600 observations BEGIN MATRICES; S FULL 1 1 FREE ! VARIANCE OF PIHATS P FULL 1 2 ! OBSERVED SCORES SIB1 SIB2 Z COMP = Z1 ! IBD 0 MATRIX N COMP = N1 ! IBD 1 MATRIX O COMP = O1 ! IBD 2 MATRIX T COMP = T1 END MATRICES; MATRIX S .5 BOUND .05 5 S 1 1 SPECIFY P SIB1 SIB2 BEGIN ALGEBRA; A = \PDFNOR(P_N_Z); ! IBD 0 PDF B = \PDFNOR(P_N_O); ! IBD 1 PDF C = \PDFNOR(P_N_T); ! IBD 2 PDF D = \SUM(((A+B+B+C)~)@(B_C)); ! EXPECTED PIHAT END ALGEBRA; MEAN D; COVARIANCE S; OPTION RSIDUAL END Summary of VL file data for group 1 P2 P1 P0 SIB1 SIB2 Code -3.0000 -2.0000 -1.0000 1.0000 2.0000 Number 600.0000 600.0000 600.0000 600.0000 600.0000 Mean 0.2683 0.4952 0.2365 0.0377 0.0283 Variance 0.0957 0.0988 0.0881 1.0675 1.0157 Minimum 0.0000 0.0000 0.0000 -2.9200 -3.0128 Maximum 1.0000 1.0000 1.0000 2.8749 3.0766 Summary of VL file data for group 2 SIB2 SIB1 PIHAT Code -2.0000 -1.0000 1.0000 Number 600.0000 600.0000 600.0000 Mean 0.0283 0.0377 0.5159 Variance 1.0157 1.0675 0.0672 Minimum -3.0128 -2.9200 0.0000 Maximum 3.0766 2.8749 1.0000 PARAMETER SPECIFICATIONS GROUP NUMBER: 1 Title QTL model with heterogeneity and weights MATRIX A This is a LOWER TRIANGULAR matrix of order 1 by 1 1 1 1 MATRIX E This is a LOWER TRIANGULAR matrix of order 1 by 1 1 1 2 MATRIX H This is a FULL matrix of order 1 by 1 It has no free parameters specified MATRIX I This is a UNIT matrix of order 3 by 2 MATRIX K This is a FULL matrix of order 3 by 1 1 1 -1 2 -2 3 -3 MATRIX M This is a FULL matrix of order 1 by 1 1 1 4 MATRIX N This is a computed FULL matrix of order 1 by 2 It has no free parameters specified MATRIX O This is a computed FULL matrix of order 2 by 2 It has no free parameters specified MATRIX Q This is a FULL matrix of order 1 by 1 1 1 3 MATRIX R This is a computed FULL matrix of order 1 by 3 It has no free parameters specified MATRIX T This is a computed FULL matrix of order 2 by 2 It has no free parameters specified MATRIX U This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX V This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX W This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX X This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX Y This is a computed FULL matrix of order 6 by 2 It has no free parameters specified MATRIX Z This is a computed FULL matrix of order 2 by 2 It has no free parameters specified GROUP NUMBER: 2 Group 2: Model pihat deviations as a function of phenotypes MATRIX A This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX B This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX C This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX D This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX N This is a computed FULL matrix of order 1 by 2 It has no free parameters specified MATRIX O This is a computed FULL matrix of order 2 by 2 It has no free parameters specified MATRIX P This is a FULL matrix of order 1 by 2 1 2 1 -1 -2 MATRIX S This is a FULL matrix of order 1 by 1 1 1 5 MATRIX T This is a computed FULL matrix of order 2 by 2 It has no free parameters specified MATRIX Z This is a computed FULL matrix of order 2 by 2 It has no free parameters specified Mx starting optimization; number of parameters = 5 MX PARAMETER ESTIMATES GROUP NUMBER: 1 Title QTL model with heterogeneity and weights Estimates of matrix R will be written to file: qtlexp.mat Format to be used is: (7(F13.6,1x)) MATRIX A This is a LOWER TRIANGULAR matrix of order 1 by 1 1 1 -2.5847E-07 MATRIX E This is a LOWER TRIANGULAR matrix of order 1 by 1 1 1 0.8096 MATRIX H This is a FULL matrix of order 1 by 1 1 1 0.5000 MATRIX I This is a UNIT matrix of order 3 by 2 MATRIX K This is a FULL matrix of order 3 by 1 1 1 0.0000 2 1.0000 3 0.0000 MATRIX M This is a FULL matrix of order 1 by 1 1 1 0.0337 MATRIX N This is a computed FULL matrix of order 1 by 2 [=M|M] 1 2 1 0.0337 0.0337 MATRIX O This is a computed FULL matrix of order 2 by 2 [=V|X_X|V] 1 2 1 1.0363 0.1904 2 0.1904 1.0363 MATRIX Q This is a FULL matrix of order 1 by 1 1 1 0.6171 MATRIX R This is a computed FULL matrix of order 1 by 3 [=A|Q|E] 1 2 3 1 -2.5847E-07 6.1712E-01 8.0962E-01 MATRIX T This is a computed FULL matrix of order 2 by 2 [=V|W_W|V] 1 2 1 1.0363 0.3808 2 0.3808 1.0363 MATRIX U This is a computed FULL matrix of order 1 by 1 [=H@A*A'] 1 1 3.3403E-14 MATRIX V This is a computed FULL matrix of order 1 by 1 [=A*A'+Q*Q'+E*E'] 1 1 1.0363 MATRIX W This is a computed FULL matrix of order 1 by 1 [=U+Q*Q'] 1 1 0.3808 MATRIX X This is a computed FULL matrix of order 1 by 1 [=U+H@Q*Q'] 1 1 0.1904 MATRIX Y This is a computed FULL matrix of order 6 by 2 [=Z_O_T] 1 2 1 1.0363E+00 3.3403E-14 2 3.3403E-14 1.0363E+00 3 1.0363E+00 1.9042E-01 4 1.9042E-01 1.0363E+00 5 1.0363E+00 3.8083E-01 6 3.8083E-01 1.0363E+00 MATRIX Z This is a computed FULL matrix of order 2 by 2 [=V|U_U|V] 1 2 1 1.0363E+00 3.3403E-14 2 3.3403E-14 1.0363E+00 Vector of OBSERVED means SIB1 SIB2 Mean 0.0377 0.0283 Vector of EXPECTED means Model number: 1 SIB1 SIB2 Mean 0.0337 0.0337 Model number: 2 SIB1 SIB2 Mean 0.0337 0.0337 Model number: 3 SIB1 SIB2 Mean 0.0337 0.0337 EXPECTED COVARIANCE MATRIX Model number: 1 SIB1 SIB2 SIB1 1.0363E+00 SIB2 3.3403E-14 1.0363E+00 Model number: 2 SIB1 SIB2 SIB1 1.0363 SIB2 0.1904 1.0363 Model number: 3 SIB1 SIB2 SIB1 1.0363 SIB2 0.3808 1.0363 Function value of this group: 3417.5020 Where the fit function is -2 * Log-likelihood of raw data GROUP NUMBER: 2 Group 2: Model pihat deviations as a function of phenotypes MATRIX A This is a computed FULL matrix of order 1 by 1 [=\PDFNOR(P_N_Z)] 1 1 0.0996 MATRIX B This is a computed FULL matrix of order 1 by 1 [=\PDFNOR(P_N_O)] 1 1 0.0994 MATRIX C This is a computed FULL matrix of order 1 by 1 [=\PDFNOR(P_N_T)] 1 1 0.0992 MATRIX D This is a computed FULL matrix of order 1 by 1 [=\SUM(((A+B+B+C)~)@(B_C))] 1 1 0.4995 MATRIX N This is a computed FULL matrix of order 1 by 2 1 2 1 0.0337 0.0337 MATRIX O This is a computed FULL matrix of order 2 by 2 1 2 1 1.0363 0.1904 2 0.1904 1.0363 MATRIX P This is a FULL matrix of order 1 by 2 1 2 1 0.0568 -0.9137 MATRIX S This is a FULL matrix of order 1 by 1 1 1 0.0650 MATRIX T This is a computed FULL matrix of order 2 by 2 1 2 1 1.0363 0.3808 2 0.3808 1.0363 MATRIX Z This is a computed FULL matrix of order 2 by 2 1 2 1 1.0363E+00 3.3403E-14 2 3.3403E-14 1.0363E+00 Vector of OBSERVED means PIHAT Mean 0.5159 Vector of EXPECTED means PIHAT Mean 0.4995 EXPECTED COVARIANCE MATRIX PIHAT PIHAT 0.0650 Function value of this group: 63.0365 Where the fit function is -2 * Log-likelihood of raw data Your model has 5 estimated parameters and 1800 Observed statistics -2 times log-likelihood of data >>> 3480.538 Degrees of freedom >>>>>>>>>>>>>>>> 1795 This problem used 10.9% of my workspace Task Time elapsed (DD:HH:MM:SS) Reading script & data 0: 0: 0: 5.84 Execution 0: 0: 0: 5.12 TOTAL 0: 0: 0:10.96 Total number of warnings issued: 0 ______________________________________________________________________________ ! Eliminate effects of QTL Multiple fit option in effect. The following MX script lines have been read: DROP @.0 3 END Summary of VL file data for group 1 P2 P1 P0 SIB1 SIB2 Code -3.0000 -2.0000 -1.0000 1.0000 2.0000 Number 600.0000 600.0000 600.0000 600.0000 600.0000 Mean 0.2683 0.4952 0.2365 0.0377 0.0283 Variance 0.0957 0.0988 0.0881 1.0675 1.0157 Minimum 0.0000 0.0000 0.0000 -2.9200 -3.0128 Maximum 1.0000 1.0000 1.0000 2.8749 3.0766 Summary of VL file data for group 2 SIB2 SIB1 PIHAT Code -2.0000 -1.0000 1.0000 Number 600.0000 600.0000 600.0000 Mean 0.0283 0.0377 0.5159 Variance 1.0157 1.0675 0.0672 Minimum -3.0128 -2.9200 0.0000 Maximum 3.0766 2.8749 1.0000 PARAMETER SPECIFICATIONS GROUP NUMBER: 1 Title QTL model with heterogeneity and weights MATRIX A This is a LOWER TRIANGULAR matrix of order 1 by 1 1 1 1 MATRIX E This is a LOWER TRIANGULAR matrix of order 1 by 1 1 1 2 MATRIX H This is a FULL matrix of order 1 by 1 It has no free parameters specified MATRIX I This is a UNIT matrix of order 3 by 2 MATRIX K This is a FULL matrix of order 3 by 1 1 1 -1 2 -2 3 -3 MATRIX M This is a FULL matrix of order 1 by 1 1 1 4 MATRIX N This is a computed FULL matrix of order 1 by 2 It has no free parameters specified MATRIX O This is a computed FULL matrix of order 2 by 2 It has no free parameters specified MATRIX Q This is a FULL matrix of order 1 by 1 It has no free parameters specified MATRIX R This is a computed FULL matrix of order 1 by 3 It has no free parameters specified MATRIX T This is a computed FULL matrix of order 2 by 2 It has no free parameters specified MATRIX U This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX V This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX W This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX X This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX Y This is a computed FULL matrix of order 6 by 2 It has no free parameters specified MATRIX Z This is a computed FULL matrix of order 2 by 2 It has no free parameters specified GROUP NUMBER: 2 Group 2: Model pihat deviations as a function of phenotypes MATRIX A This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX B This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX C This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX D This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX N This is a computed FULL matrix of order 1 by 2 It has no free parameters specified MATRIX O This is a computed FULL matrix of order 2 by 2 It has no free parameters specified MATRIX P This is a FULL matrix of order 1 by 2 1 2 1 -1 -2 MATRIX S This is a FULL matrix of order 1 by 1 1 1 5 MATRIX T This is a computed FULL matrix of order 2 by 2 It has no free parameters specified MATRIX Z This is a computed FULL matrix of order 2 by 2 It has no free parameters specified Mx starting optimization; number of parameters = 4 MX PARAMETER ESTIMATES GROUP NUMBER: 1 Title QTL model with heterogeneity and weights Estimates of matrix R will be written to file: qtlexp.mat Format to be used is: (7(F13.6,1x)) MATRIX A This is a LOWER TRIANGULAR matrix of order 1 by 1 1 1 0.6272 MATRIX E This is a LOWER TRIANGULAR matrix of order 1 by 1 1 1 0.8051 MATRIX H This is a FULL matrix of order 1 by 1 1 1 0.5000 MATRIX I This is a UNIT matrix of order 3 by 2 MATRIX K This is a FULL matrix of order 3 by 1 1 1 0.0000 2 1.0000 3 0.0000 MATRIX M This is a FULL matrix of order 1 by 1 1 1 0.0330 MATRIX N This is a computed FULL matrix of order 1 by 2 [=M|M] 1 2 1 0.0330 0.0330 MATRIX O This is a computed FULL matrix of order 2 by 2 [=V|X_X|V] 1 2 1 1.0416 0.1967 2 0.1967 1.0416 MATRIX Q This is a FULL matrix of order 1 by 1 1 1 0.0000 MATRIX R This is a computed FULL matrix of order 1 by 3 [=A|Q|E] 1 2 3 1 0.6272 0.0000 0.8051 MATRIX T This is a computed FULL matrix of order 2 by 2 [=V|W_W|V] 1 2 1 1.0416 0.1967 2 0.1967 1.0416 MATRIX U This is a computed FULL matrix of order 1 by 1 [=H@A*A'] 1 1 0.1967 MATRIX V This is a computed FULL matrix of order 1 by 1 [=A*A'+Q*Q'+E*E'] 1 1 1.0416 MATRIX W This is a computed FULL matrix of order 1 by 1 [=U+Q*Q'] 1 1 0.1967 MATRIX X This is a computed FULL matrix of order 1 by 1 [=U+H@Q*Q'] 1 1 0.1967 MATRIX Y This is a computed FULL matrix of order 6 by 2 [=Z_O_T] 1 2 1 1.0416 0.1967 2 0.1967 1.0416 3 1.0416 0.1967 4 0.1967 1.0416 5 1.0416 0.1967 6 0.1967 1.0416 MATRIX Z This is a computed FULL matrix of order 2 by 2 [=V|U_U|V] 1 2 1 1.0416 0.1967 2 0.1967 1.0416 Vector of OBSERVED means SIB1 SIB2 Mean 0.0377 0.0283 Vector of EXPECTED means Model number: 1 SIB1 SIB2 Mean 0.0330 0.0330 Model number: 2 SIB1 SIB2 Mean 0.0330 0.0330 Model number: 3 SIB1 SIB2 Mean 0.0330 0.0330 EXPECTED COVARIANCE MATRIX Model number: 1 SIB1 SIB2 SIB1 1.0416 SIB2 0.1967 1.0416 Model number: 2 SIB1 SIB2 SIB1 1.0416 SIB2 0.1967 1.0416 Model number: 3 SIB1 SIB2 SIB1 1.0416 SIB2 0.1967 1.0416 Function value of this group: 3432.6140 Where the fit function is -2 * Log-likelihood of raw data GROUP NUMBER: 2 Group 2: Model pihat deviations as a function of phenotypes MATRIX A This is a computed FULL matrix of order 1 by 1 [=\PDFNOR(P_N_Z)] 1 1 0.0992 MATRIX B This is a computed FULL matrix of order 1 by 1 [=\PDFNOR(P_N_O)] 1 1 0.0992 MATRIX C This is a computed FULL matrix of order 1 by 1 [=\PDFNOR(P_N_T)] 1 1 0.0992 MATRIX D This is a computed FULL matrix of order 1 by 1 [=\SUM(((A+B+B+C)~)@(B_C))] 1 1 0.5000 MATRIX N This is a computed FULL matrix of order 1 by 2 1 2 1 0.0330 0.0330 MATRIX O This is a computed FULL matrix of order 2 by 2 1 2 1 1.0416 0.1967 2 0.1967 1.0416 MATRIX P This is a FULL matrix of order 1 by 2 1 2 1 0.0568 -0.9137 MATRIX S This is a FULL matrix of order 1 by 1 1 1 0.0675 MATRIX T This is a computed FULL matrix of order 2 by 2 1 2 1 1.0416 0.1967 2 0.1967 1.0416 MATRIX Z This is a computed FULL matrix of order 2 by 2 1 2 1 1.0416 0.1967 2 0.1967 1.0416 Vector of OBSERVED means PIHAT Mean 0.5159 Vector of EXPECTED means PIHAT Mean 0.5000 EXPECTED COVARIANCE MATRIX PIHAT PIHAT 0.0675 Function value of this group: 85.0880 Where the fit function is -2 * Log-likelihood of raw data Your model has 4 estimated parameters and 1800 Observed statistics -2 times log-likelihood of data >>> 3517.702 Degrees of freedom >>>>>>>>>>>>>>>> 1796 Saturated model fit* >>>>>>>>>>> 3480.538 Saturated model df* >>>>>>>>>>> 1795 Difference Chi-squared >>>>>>>> 37.163 Difference d.f. >>>>>>>>>>>>>>> 1 Probability >>>>>>>>>>>>>>>>>>>> 0.000 Akaike's Information Criterion > 35.163 * Saturated model statistic computed earlier in this job This problem used 10.9% of my workspace Task Time elapsed (DD:HH:MM:SS) Reading script & data 0: 0: 0: 0.00 Execution 0: 0: 0: 9.83 TOTAL 0: 0: 0: 9.83 Total number of warnings issued: 0 ______________________________________________________________________________