ESTIMATION OF BREEDING VALUES AND THEIR ACCURACIES USING MULTIVARIATES ANIMAL MODEL ANALYSIS FOR GROWTH TRAITS IN THREE LOCAL STRAINS OF CHICKENS

Transcription

1 Egypt. Poult. Sci. Vol. 0 (IV) Dec. 000 (98-00) ESTIMATION OF BREEDING VALUES AND THEIR ACCURACIES USING MULTIVARIATES ANIMAL MODEL ANALYSIS FOR GROWTH TRAITS IN THREE LOCAL STRAINS OF CHICKENS M. M. IRAQI, A.F.M. EL-LABBAN * AND M.H. KHALIL ** Deprtment Of Animl Production At Moshtohor, Zgzig University, Egypt * Animl Production Reserch Institute, Ministry of Agriculture, Dokki, Ciro, Egypt ** College of Agriculture nd Veterinry Medicine, King Sud University, Sudi Arbi (Received in 6/0/000) (Accepted in 8//000) ABSTRACT: Breeding vlues nd their ccurcies were estimted using multi-trits Animl Model for birds with nd without records of three locl strins of chickens. A totl number of 96, 37 nd 030 chicks from Golden Montzh (GM), Silver Montzh (SM) nd Mtrouh (MA) strins, respectively, were used. Dt of individul body weight t htch (BW0), 4(BW4), 8 (BW8) nd (BW) weeks of ge were collected from breeding experiment of one genertion. Dily gin (DG) trits between intervls of htch-4 (DG4), 4-8 (DG8) nd 8- (DG) weeks were lso used. Results show tht estimtes of heritbility ( h ) in GM nd SM strins for BW t htch were higher thn tht t weeks. The estimtes re 0.45 nd 0. for BW0 nd 0. nd 0.8 for BW in GM nd SM strins, respectively. For birds with records, the rnges in estimtes of predicted breeding vlue (PBV) in GM, SM nd MA strins were 4.6, 4.7 nd 4.0 grms for BW0, 3.5, 8.7 nd 5.9 grms for BW4, 37.9,. nd 68. grms for BW8 nd.8, 78. nd 88.8 grms for BW, respectively. The respective rnges in estimtes of PBV for DG trits were.5, 0.58 nd.8 grms for DG4,.99, 4.05 nd 3.3 grms for DG8 nd 9.54, 4.3 nd 8.73 grms for DG. For sires nd dms without records in GM, SM nd MA strins, the minimum nd mximum estimtes of breeding vlues predicted for sires (PBVS) nd dms (PBVD) of birds nd their rnges for BW nd DG trits hd the sme trend obtined for birds with records. MA strin recorded the highest rnges of PBV, PBVS nd PBVD for most growth trits compred to those recorded by GM nd SM strins. Accurcies of minimum nd mximum estimtes of sires breeding vlues for

2 M. M. Irqi, et l. BW nd DG trits predicted for MA strin were the highest followed by SM nd GM strins. Also, ccurcies of predicted breeding vlues for dms of birds recorded by MA nd SM strins were higher thn those recorded by GM strin. The verges of ccurcies in estimtes of predicted breeding vlues for sires of birds in the three strins were nerly the sme. INTRODUCTION Growth performnce in poultry should be selected on the bsis of their estimted breeding vlues rther thn on phenotypic performnce. The ccurcy of estimtion of the breeding vlue of the chicks becomes more precise together with n extending for the informtion not only vilble on their own performnce test but lso on both the full nd hlf sibs s well s of the ncestors. The Animl Model is nowdys pplied in some species of poultry, but it hs not been widely used. Derivtive-Free Restricted Mximum Likelihood Algorithm (DFREML) llows unified pproch for genetic evlution nd estimtion of vrince components for chickens (Mielenz et l., 994). Recently, multivrite predictions of the genetic merit of the individul (e.g. MTDFREML, MTGSAM, PEST, etc.) requires specifiction of the covrince structure mong the trits considered for true or conceptul bse genertion (this is neglected with single-trit nlysis). In Egypt, Irqi (999) crried out n ttempt bsed on comprison between single- nd multi-trits Animl Models for growth trits in Dokki-4 chickens. He concluded tht pplying multi-trits Animl Model in evlution llows estimtion of dditive genetic vrince without bis, consequently the predictors re BLUP ssocited with lower predicted error vrince (PEV) nd n increse in selection efficiency. An erly ttempt ws done by Abdelltif nd Abdou (977) to estimte breeding vlues of Fyoumi cocks using different methods (Hungrin Stndrd, Luprecht nd Siler-Vchl) of progeny testing. For other locl strins like Silver nd Golden Montzh, Mtrouh, Bndr, Alexndri, etc, no ttempts hve been mde to predict the breeding vlues of the birds using new methodology such s single or multi-trits Animl Models. Therefore, this study ws imed: () to estimte the dditive genetic vrince nd heretbility for growth trits of three locl strins (e.g. Silver Montzh, Golden Montzh nd Mtrouh chickens), () to predict the breeding vlues for sires, dms nd their progenies nd (3) to estimte the ccurcy nd stndrd errors of the predicted estimtes. 98

3 Chickens, Egyptin locl strins, Growth performnce, heritbility, Breeding vlues, Multi-trits Animl Model. MATERIAL AND METHODS This experiment ws crried out t Inshs Poultry Breeding Reserch Sttion, Animl Production Reserch Institute, Ministry of Agriculture, Egypt. Three pedigreed locl strins of poultry nmed Golden Montzh (GM), Silver Montzh (SM) nd Mtrouh (MA) were used. The SM strin ws developed from the crossing of Rhode Islnd Red mles (exotic breed) nd Dokki-4 femles (ntive breed) for three consecutive genertions together with selection, then intense mtings nd selection for severl genertions were prcticed (Mhmoud et l., 974b). One GM strin ws developed from the crossing of Rhode Islnd Red mles nd Dokki-4 femles for five consecutive genertions together with selection, then intense mtings nd selection for severl genertions were prcticed (Mhmoud et l., 974c). The MA strin ws developed from the crossing of White Leghorn mles nd Dokki-4 femles for three consecutive genertions together with selection, then intense mtings nd selection for severl genertions were prcticed (Mhmoud et l., 974). The experimentl work ws crried out for one genertion strted in 997. A totl number of 5, 6 nd 7 pedigreed cocks nd 4, 33 nd 34 pedigreed pullets were chosen rndomly t sexul mturity from bse popultions of GM, SM nd MA strins, respectively, to be the prents of the first genertion. The number of progeny produced from the three strins were 96, 37 nd 030 chicks, respectively. Chicks were fed d libitum during the growing period on commercil rtion contining % crude protein, 3.9% crude ft nd 3.7% crude fibers. All chicks of one-dy old were wing-bnded nd rered in floor brooder, then trnsferred to the rering houses. All birds were treted nd medicted similrly throughout the experimentl period nd they were rised under the sme mngeril nd climtic conditions s possible. Dt Dt of body weight (grm) t htch (BW0), 4 weeks (BW4), 8 weeks (BW8), nd weeks (BW) of ge were collected on ll individuls of the first genertion. Dily gins in weight (DG) between intervls of 0-4 (DG4), 4-8 (DG8) nd 8- (DG) weeks were lso computed. Model of nlysis Dt of ech strin were nlysed seprtely using multi-trits Animl Model (Boldmn et l, 995). The model of multivrite nlysis resembles stck of the single-trit Animl Model for ech trit. The model of multivrite nlysis for two trits could be written s: 983

4 M. M. Irqi, et l. 984 Where y i = Vector of the i th trit observed on birds; b i = Vector of fixed effect of sex for the i th trit; u i = Vector of rndom bird effects for the i th trit; X i nd Z i re incidence mtrices relting records of the i th trit to fixed effects nd the rndom bird effects, respectively. Writing out the equtions for ech trit seprtely in the Multi-trits Animl Model, the mixed model equtions (MME) become: Where A = inverse of the numertor reltionship mtrix mong birds. In the present study, the MME in multi-trits Animl Model re being too lrge when we hve more thn two trits (e.g. 4 trits for body weights nd 3 trits for dily gins). The multivrite Animl Model is considering the reltionship coefficient mtrix (A - ) mong birds in estimtion (Korhonen, 996). All clcultions of BLUP estimtes for multi-trits Animl Model were crried out using the MTDFREML progrm (Boldmn et l., 995) dpted to use the sprse mtrix pckge, SPARSPAK (George nd Ng, 984). Convergence ws ssumed when the vrince of the log-likelihood vlues in the simplex reched <0-6. Occurrence of locl mximl ws checked by repetedly restrting the nlyses until the log-likelihood did not chnge beyond the first deciml. Estimtes of vrince, covrince nd heritbility A multi-trits Animl Model ws used to estimte direct dditive genetic, error, phenotypic vrinces nd heritbility. Vrinces nd covrinces obtined by the sire model (REML method using procedure VARCOMP, SAS, 996) were used s strting vlues (guessed vlues) for y y Z 0 0 X 0 0 e e u u Z b b X y y y y y X' y X' y X' y X' ˆ ˆ ˆ ˆ ' u u b b A Z A Z X X A Z A Z X X Z X' Z X' X X' X X' Z X' Z X X X' X X'

5 Chickens, Egyptin locl strins, Growth performnce, heritbility, Breeding vlues, Multi-trits Animl Model. the estimtion of vrince nd covrince components using multi-trits Animl Model. Heritbility ws computed ccording to Boldmn et l. (995) s: h e Where nd e re vrinces due to effects of direct dditive genetic nd rndom error, respectively. Estimtion of breeding vlues The vrinces nd covrinces estimted by multi-trits Animl Model nlysis were used for estimtion of predicted breeding vlues (PBV), stndrd errors nd their ccurcies of predictions ( r AA ). Solutions for equtions of birds were computed from the pedigree file, one bird t time for birds with records nd birds without records, i.e. sire nd dms. A digonl element (d t ) nd n djusted right-hnd side ( y* t) were ccumulted with ech pedigree file record for the t th bird. For birds with nd without records, the formul (Kennedy, 989) used to estimte the predicted breeding vlues (PBV) ws: PBV = [ y* t/d t ]; where y* t/d t = breeding vlues of the birds with nd without records. The ccurcy of predicted breeding vlue for ech individul ws estimted ccording to Henderson (975) s: + F - d j r AA j Where r AA = the ccurcy of prediction of the i th bird s breeding vlue for birds with nd without records; F j = inbreeding coefficient of birds (which equl to zero s clculted using MTDFREML progrm of Boldmn et l., 995); d j = the j th digonl element of inverse of the pproprite block coefficient mtrix; nd = e/. Stndrd error (SE) of predicted breeding vlue for ech individul ws estimted s follows: s. e. = Where d j nd e were defined bove. d p j e RESULTS AND DISCUSSION Mens nd vritions Mens, phenotypic stndrd devitions (SD) nd percentges of vribility of body weights (BW) nd dily gins (DG) in GM, SM nd MA 985

6 M. M. Irqi, et l. strins re given in Tble. Results show tht GM nd SM strins hd higher in growth trits thn MA strin. Mens in the present study were generlly higher thn recorded by Mhmoud et l. (974&b&c) for the sme strins. On the other hnd, DG trits in GM nd SM strins re high compred to in MA strin. This my be due to the GM nd SM strins originted from crossing Rhode Islnd Red (s dul purpose breed) mles with Dokki-4 femles, while MA strin originted from crossing White Leghorn (s egg-type breed) mles with Dokki-4 femles (Mhmoud et l., 974&b&c). Percentges of vribility for ech trit of BW nd DG in the three strins tend to increse with the increse of the ge. The sme trend ws shown by Sbr (990), Khlil et l. (993) nd Irqi (999). Estimtes of V% for most of growth trits in MA strin were higher thn those in SM nd GM trins. The percentge of vribility rnged from 8.38 to 9.03%, 6.68 to 4.34% nd 8.9 to 3.38% for growth trits in MA, SM nd GM strins, respectively. Vrince components nd heritbilities Estimtes of dditive ( ), error ( e ), phenotypic ( p ) nd heritbility ( h ) for growth trits in GM, SM nd MA strins re presented in Tble. Results show tht most percentges of for growth trits in MA strin were higher thn in SM nd GM strins. These results were generlly within the rnge of those estimtes obtined by Irqi et l. (000) in the comprtive study mong different methods to estimtes vrince component for the sme strins. Commonly, estimtes of in MA nd GM strins in the present study were somewht high compred to findings of Irqi (999) on Dokki-4 chickens using single- nd multi-trits Animl Model. This might be due to these strins were not subjected to ny intensive progrm of selection (Irqi et l., 000). Estimtes of heritbility ( h ) for growth trits in the three strins showed the similr trend of percentges of dditive genetic vrince (Tble ). Estimtes of h in GM nd SM strins for BW t htch were higher thn t lter ges (t weeks). The estimtes re 0.45 nd 0. for BW0 nd 0. nd 0.8 for BW in GM nd SM strins, respectively. This might be due to the smll mternl effects, i.e. decresing the non-dditive genetic vrince effects (Dnbro et l., 995 nd Irqi 999). Thus, we would recommend the Egyptin poultry breeder to select these strins t erly ges without witing to lter ges to sve time nd efforts. Estimtes 986

7 Chickens, Egyptin locl strins, Growth performnce, heritbility, Breeding vlues, Multi-trits Animl Model. of h in the present study were generlly within the rnge of those estimtes obtined for the sme strins by Sbr (990) nd Irqi et l. (000). Breeding vlues estimted for birds with records The minimum nd mximum estimtes of predicted breeding vlue (PBV) nd their rnges (i.e. the difference between the mximum nd minimum vlue), stndrd errors (SE) nd ccurcy of ech predictor ( r AA ) for growth trits of ll birds nd the top 50% of progeny birds to be selected in GM, SM nd MA strins re presented in Tble 3. For the list of ll birds, the rnges in estimtes of PBV in GM, SM nd MA strins were 4.6, 4.7 nd 4.0 grms for BW0, 3.5, 8.7 nd 5.9 grms for BW4, 37.9,. nd 68. grms for BW8 nd.8, 78. nd 88.8 grms for BW, respectively. The respective rnges in estimtes of PBV for DG trits were.5, 0.58 nd.8 grms for DG4,.99, 4.05 nd 3.3 grms for DG8 nd 9.54, 4.3 nd 8.73 grms for DG. Rnges mentioned before indicte tht MA strin recorded higher rnges in estimtes of PBV for BW nd DG trits thn the GM nd SM strins. As stted before, GM nd SM strins originted from crossing Rhode Islnd Red mles with Dokki-4 femles, while MA strin originted from crossing White Leghorn mles with Dokki- 4 femles (Mhmoud, 974&b&c). Consequently, the three strins originted from one dm-breed (Dokki-4) while they differed in sire-breed in terms of Rhode Islnd Red for both GM nd SM nd White Leghorn for MA strin. The Rhode Islnd Red is dul-purpose breed nd the White Leghorn is n egg-type breed nd this my explin why genetic vrition for growth performnce in MA strin could be high compred with SM nd GM. The high estimtes of PBV for growth trits in MA strin indicte tht improvement of growth performnce of this strin could be chieved through selection reltive to the other two strins studied. Accurcies ( r AA ) of minimum nd mximum estimtes of PBV recorded for BW nd DG trits of birds with records were moderte to high in most cses (Tble 3). The estimtes of r AA for BW trits rnged from 0.46 to 0.65 in GM strin, 0.56 to 0.68 in SM nd 0.58 to 0.69 in MA, while the respective figures for DG trits rnged from 0.6 to 0.73, 0.65 to 0.66 nd 0.66 to 0.73 (Tble 3). However, the ccurcies in PBV recorded by MA nd SM strins were high compred to estimtes obtined in GM. This my due to the smll number of records used in GM (96 records) vs 37 records used in SM nd 030 records used in MA. Pribyl nd Pribylov (99) found tht relibility of PBV ws 0.60 for body weight of birds with records t 0-week of ge. Bsed on multi-trits Animl Model, Irqi (999) 987

8 M. M. Irqi, et l. found tht verges of ccurcies for estimtes of PBV for BW nd DG trits of Dokki-4 birds with records were 0.66 nd 0.65, respectively. The number of birds with positive estimtes of PBV nd their percentges (Tble 4) indicte tht MA strin recorded high percentges (55%) of birds with positive signs for growth trits compred to those estimtes recorded by GM nd SM strins. Percentges of birds with positive signs in the present study for ll strins were higher thn those findings recorded by Irqi (999) for Dokki-4 chickens (4%). This indictes tht GM, SM nd MA strins hve high PBV with positive signs, nd this leds to stte tht the top 50% of birds to be selected ll hd positive PBV. Thus, selection of birds themselves could be more effective method (individul selection) to improve growth trits of the three strins of chickens under locl conditions. Breeding vlues estimted for sires of birds (birds without records) The minimum nd mximum estimtes of breeding vlues predicted for sires of birds (PBVS) nd their rnges for BW nd DG trits in GM, SM nd MA strins (Tble 5) indicte tht estimtes for birds without records hd the sme trend obtined for birds with records. The rnges in estimtes of PBVS in GM, SM nd MA strins were 3.3, 4. nd 3.3 grms for BW0,.8, 5.4 nd 3.6 grms for BW4, 37.6, 8.0 nd 6.7 grms for BW8 nd 83.5, 76.5 nd 39.9 grms for BW, respectively. The corresponding rnges in estimtes of PBVS for DG trits were 0.8, 0. nd. grms for DG4,.8,. nd.3 grms for DG8, nd 4.5,. nd 6.9 grms for DG. Rnges given before indicte tht MA strin recorded the highest estimtes of PBVS for ll growth trits compred to those recorded by GM nd SM strins. In greement with the present results, Irqi (999) found tht the predicted sire trnsmitting bilities for Dokki-4 chickens were 0.09, 0.7, 4.84 nd.43 grms for BW0, BW4, BW8, BW, respectively, nd 0.35, 0.7 nd 0.03 grms for DG4, DG8 nd DG, respectively. From one ge to nother, the ccurcy ( r AA ) of minimum nd mximum estimtes of PBVS for growth trits of sires of birds in the three strins were high nd rnged from 0.68 to 0.90 in GM, 0.74 to 0.85 in SM nd 0.75 to 0.86 in MA (Tble 5). The pttern of chnge of these estimtes hd the sme trend obtined for birds with records (Tble 3). The estimtes for GM, SM nd MA strins verged 0.76, 0.8 nd 0.80 for BW trits nd verged 0.86, 0.83 nd 0.83 for DG trits, respectively. Across ll minimum nd mximum estimtes of PBVS, the verges of ccurcies in ll the three locl strins were nerly the sme (Tble 5). This my be due to the numbers of sires used in the three strins were nerly equl (i.e. 5, 6 988

9 Chickens, Egyptin locl strins, Growth performnce, heritbility, Breeding vlues, Multi-trits Animl Model. nd 7 sires were used from GM, SM nd MA strins, respectively). Pribyl nd Pribylov (99) found tht relibility of PBVS ws 0.84 for body weights of sires of birds t 0-week of ge. Recently, Irqi (999) found tht verges of ccurcy cross ll the minimum nd mximum estimtes of PBVS were 0.8 nd 0.75 for BW nd DG trits in Dokki-4 chickens, respectively. The ccurcy in PBVS estimted for sires without records (PBVS) were higher thn those PBV estimted for birds with records (Tbles 3&5). This my be due to tht ech sire hd n verge of 67 progeny records, while birds with records hd just only one record in estimtion of breeding vlue. Korhonen (996) reported tht the heritbility of the trit nd the mount of informtion utilized in evlution could ffect the relibility of the predictors. The number of sires with positive estimtes of PBVS nd their percentges for most growth trits (Tble 6) indicte tht predictors recorded by sires of MA strin, s stted before for PBVS, were mostly higher thn those recorded by GM nd SM strins. Similrly, Irqi (999) found tht verges of percentges for PBVS with positive sign were 48.5% nd 50% for BW nd DG trits, respectively. Breeding vlues estimted for dms (PBVD) of birds (birds without records) The minimum nd mximum estimtes of PBVD nd their rnges (Tble 7) indicte tht these estimtes hd the sme trend obtined for birds with records (Tble 3). The rnges in estimtes of PBVD in GM, SM nd MA strins were.7, 3. nd 3.0 grms for BW0, 4.6, 3.0 nd 37.9 grms for BW4,., 8.8 nd 56. grms for BW8 nd 80.3, 4.9 nd 68.0 grms for BW, respectively. The corresponding rnges in estimtes of PBVD for DG trits were 0.53, 0.48 nd. grms for DG4,.33,.64 nd 3.96 grms for DG8 nd 4.45, 3.9 nd 4.8 grms for DG; i.e. MA strin recorded the highest estimtes of PBVD for most growth trits. Similr to the present results, Irqi (999) with Dokki-4 chickens reported tht the predicted trnsmitting bilities for dms were 0.55, 4.3, 9.7 nd.77 grms for BW0, BW4, BW8 nd BW, respectively, nd 0.0, 0.47 nd 0.33 grms for DG4, DG8 nd DG, respectively. Across ll minimum nd mximum estimtes of PBVD for growth trits of dms of birds, the ccurcies of these estimtes in GM, SM nd MA strins were moderte to high (Tble 7). The estimtes of r AA in GM, SM nd MA strins verged 0.50, 0.57 nd 0.56 for body weight trits nd 0.38, 0.57 nd 0.63 for gin in weight trits, respectively. These results 989

10 M. M. Irqi, et l. indicte tht ccurcy in estimtes of PBVD for dms of MA nd SM strins were higher thn those recorded for GM. This my be due to the fct tht numbers of dms used in evlution for MA nd SM were the sme (i.e. 33 nd 34 dms) irrespective to 4 dms used in GM strin (Korhonen, 996). On the other hnd, the ccurcy in estimtes of PBVD recorded by the dms of birds without records were lower thn those recorded by the birds themselves (birds with records) nd their sires without records (Tbles 3&5&7). This my be due to smll numbers of progeny per dm utilized (ech dm hd n verge of 8 progeny records). The numbers of dms nd their percentges hving positive estimtes of PBVD (Tble 8) indicte tht percentges of PBVD with positive signs for ll growth trits were nerly verged 50% in the three locl strins. CONCLUSIONS () Estimtes of h in GM nd SM strins for BW trits t htch were higher thn t lter ges (t weeks). Therefore, we would recommend the Egyptin poultry breeder to select these strins t erly ges without witing to lter ges to sve time nd efforts. () For birds with nd without records, MA strin recorded higher rnges in estimtes of PBV, PBVS nd PBVD for BW nd DG trits thn those recorded by GM nd SM strins. It could be firly, therefore, to stte tht using of MA strin could be n encourging fctor for the poultry breeder to involve this strin in ny improvement progrm to improve growth performnce of locl chickens in Egypt. (3) Accurcy of PBV incresed s the number of records incresed. This ws evidenced in this study since ccurcy in PBV for dms (with n verge of 8 progeny records) were lower thn ccurcy in PBV for sires (with n verge of 67 progeny records). 990

11 Egypt. Poult. Sci. Vol. 0 (IV) Dec. 000 (98-00) Tble. Mens nd coefficients of vrition (V%) for body weight (BW) nd dily gins (DG) in Golden Montzh, Silver Montzh nd Mtrouh strins. Golden Montzh Silver Montzh Mtrouh Trit Symble No. M±SD V% No. M±SD V% No. M±SD V% Body weights (g): BW0 Body weight t htch ± ± ± BW4 BW0 Body weight t 4 weeks ± ± ± BW8 Body weight t 8 weeks ± ± ± BW Body weight t week ± ± ± Dily gins (g): DG4 Dily gins from htch-4 weeks ± ± ± DG8 Dily gins from 4-8 weeks ± ± ± DG Dily gins from 8- week ± ± ±

12 M. M. Irqi, et l. Tble. Estimtes of direct dditive genetic ( ), error ( e ), phenotypic ( p ) vrinces nd heritbility ( h ) for growth trits in in Golden Montzh, Silver Montzh nd Mtrouh strins. Trit+ Golden Montzh Silver Montzh Mtrouh % Body weight (BW): e % p h % e % p h % e % BW BW BW BW Dily gins (DG): DG DG DG Trits s defined in Tble. p h 98

13 Chickens, Egyptin locl strins, Growth performnce, heritbility, Breeding vlues, Multi-trits Animl Model. Tble 3. Minimum, mximum nd rnges of predicted breeding vlues (PBV) for birds with records, their stndrd errors (SE) nd ccurcy of prediction ( r AA ) estimted by multi-trits niml model for growth trits in Golden Montzh, Silver Montzh nd Mtrouh strins *. Trit + Rn Rng Golden Montzh Silver Montzh Mtrouh Minimum Mximum in PBV Minimum Mximum in PBV Minimum Mximum PB SE r A PB SE r A PB SE r A PB SE r A PBV SE r A PBV SE r A Body weights in grms (BW): BW BW BW BW Dily gins in grms (DG): DG DG DG Trits s defined in Tble. * Totl numbers of progeny with records evluted were 96, 37 nd 030 individuls for Golden Montzh, Silver Montzh nd Mtrouh strins, respectively. Rng in PBV 983

14 M. M. Irqi, et l. Tble 4. Numbers (nd percentges) of birds with records which hving positive estimtes of breeding vlues predicted by multi-trits niml model for growth trits in Golden Montzh (GM), Silver Montzh (SM) nd Mtrouh (MA) strins. Trit Birds with positive PBV + Body weights (BW): GM No % SM No % MA No % BW BW BW BW Averge Dily gins (DG): DG DG DG Averge Totl numbers of birds evluted were 96, 37 nd 030 individuls in GM, SM nd MA strins, respectively. 984

15 Chickens, Egyptin locl strins, Growth performnce, heritbility, Breeding vlues, Multi-trits Animl Model. Tble 5. Minimum, mximum nd rnges of predicted breeding vlues for sires without records (PBVS), their stndrd errors (SE) nd ccurcy of prediction ( r AA ) estimted by multi-trits niml model for growth trits in Golden Montzh, Silver Montzh nd Mtrouh strins +. Golden Montzh Silver Montzh Mtrouh Trit Rnge Rnge Minimum Mximum in PBVS Minimum Mximum in PBVS Minimum Mximum PBVS SE r AA PBVS SE r AA PBVS SE r AA PBVS SE r AA PBVS SE r AA PBVS SE r AA Rnge in PBVS Body weights in grms (BW): BW BW BW BW Dily gins in grms (DG): DG DG DG Totl numbers of birds without records evluted were 5, 6 nd 7 sires for Golden Montzh, Silver Montzh nd Mtrouh strins, respectively. 985

16 M. M. Irqi, et l. Trit Tble 6. Numbers (nd percentges) of sires without records which hving positive estimtes of breeding vlues (PBVS) predicted by multi-trits niml model for growth trits in Golden Montzh (GM), Silver Montzh (SM) nd Mtrouh (MA) strins. Birds with positive PBVS + Body weights (BW): GM No % SM No % MA No % BW BW BW BW Averge Dily gins (DG): DG DG DG Averge Totl numbers of sires evluted were 5, 6 nd 7 in GM, SM nd MA strins, respectively. 986

17 Chickens, Egyptin locl strins, Growth performnce, heritbility, Breeding vlues, Multi-trits Animl Model. Tble 7. Minimum, mximum nd rnges of predicted breeding vlues for dms without records (PBVD), their stndrd errors (SE) nd ccurcy of prediction ( r AA ) estimted by multi-trits niml model for growth trits in Golden Montzh, Silver Montzh nd Mtrouh strins +. Golden Montzh Silver Montzh Mtrouh Trit Minimum Mximum Rnge Minimum Mximum Rnge Minimum Mximum Rnge in in in SE SE PBVD PBVD PBVD PBVD SE r AA PBVD SE r AA Body weights in grms (BW): PBVD SE r AA PBVD SE r AA PBV D r AA PBV D BW BW BW BW Dily gins in grms (DG): DG DG DG Totl numbers of birds without records evluted were 4, 33 nd 34 dms in Golden Montzh, Silver Montzh nd Mtrouh strins, respectively. r AA 987

18 M. M. Irqi, et l. Tble 8. Numbers (nd percentges) of dms without records which hving positive estimtes of breeding vlues (PBVD) predicted by multi-trits niml model for growth trits in Golden Montzh (GM), Silver Montzh (SM) nd Mtrouh (MA) strins +. Birds with positive PBVD + Trit Body weights (BW): GM No % SM No % MA No % BW BW BW BW Averge Dily gins (DG): DG DG DG Averge Totl numbers of dms evluted were 4, 33 nd 34 in GM, SM nd MA strins, respectively. 988

19 Egypt. Poult. Sci. Vol. 0 (IV) Dec. 000 (98-00) REFERENCES Abdelltif, M. nd Abdou, F. H. (977). Using different methods of progeny testing for breeding vlues estimtion of Fyoumi cocks. Act Agronomic Acdem. Scient. Hungrice 6: Boldmn, K. G., Kriese, L. A., Vn Vleck, L. D., Vn Tssell, C. P. & Kchmn, S. D. (995). A mnul for use of MTDFREML. A set of progrms to obtin estimtes of vrinces nd covrinces [DRAFT]. U.S. Deprtment of Agriculture, Agriculturl Reserch Service, USA. George, A. & Ng, E. (984). A new relese of SPARSPAK: The wterloo sprse mtrix pckge. Mimeo, Dept. Computer Sci., Univ. Wterloo, Wterloo, ON., Cnd (Agri. Res. Serv., USA, 995). Henderson, C. R. (975). Best liner unbised estimtion nd prediction under selection model. Biometrics 3(): Irqi, M. M. (999). Estimtion nd evlution of sire trnsmitting bilities for growth trits in chickens. Ph.D. Thesis, Fculty of Agriculture, Moshtohor, Zgzig University, Egypt. Irqi, M. M., El-Lbbn, A. F. M. nd Afifi, E.A. (000). A comprison of four methods of estimting vrince components nd heritbility for growth trits in chickens. Proc. Conf. Anim. Prod. In The th Century, Skh, 8-0 April 000: 345:358. Khlil, M. H., Hnfi, M., El-Lbbn, A. E. F., nd Irqi, M. M. (993). Genetic evlution of growth trits in Dokki-4 chickens. Egyptin J. Anim. Prod. 30(): Kennedy, B. W. (989). ANIMAL MODEL BLUP - Crsmus intensive grdute course. University of Guelph, Dublin. Korhonen, T. (996). "The diry cttle evlution of 996" (h: 4:3) (Article). Mhmoud, T.H., Mdkour, Y.H., Syed, I.F. & Hrirh, K.M. (974). " Mtrouh" - A new breed of chickens. Agriculture Reserch Review, Egyptin 5(6): Mhmoud, T.H., Syed, I.F. & Mdkour, Y.H. (974b). " The Silver Montzh" - A new vriety of chickens. Agriculture Reserch Review, Egyptin 5(6): Mhmoud, T.H., Syed, I.F. & Mdkour, Y.H. (974c). " The Golden Montzh" - A new vriety of chickens. Agriculture Reserch Review, Egyptin 5(7): 5-60.

20 M. M. Irqi, et l. Mielenz, N., Groeneveld, E., Muller, J. & Spilke, J. (994). Simultneous estimtion of vrinces nd covrinces using REML nd Henderson 3 in selected popultion of White Leghorns. British Poultry Science 35(5): Pribyl, J. nd Pribylov, J. (99). The use of BLUP method for the construction of selection indexes in egg-lying poultry. Scienti Agriculture Bohemoslovc UVTIZ 3(): Sbr, Z. E. (990). Estimtion of heterosis nd combining bilities for some economic trits in chickens. M. Sc. Thesis, Fculty of Agriculture, Moshtohor, Zgzig University, Egypt SAS. (996). SAS' Procedure Guide. "Version 6. Ed." SAS Institute Inc., Cry, NC, USA. 98

21 Chickens, Egyptin locl strins, Growth performnce, heritbility, Breeding vlues, Multi-trits Animl Model. الهمخص العرتى ثقدير القين الثرتوية وهعبهالت الدقة تبسثخدان ثحميل ىهوذخ الحيواو الوراحى هثعدد الصفبت - لصفبت الىهو فى حالج سالالت هحمية هو الدذبخ هحهود هغرتى عراقى * عتد الفثبر هحهد المتبو ** هبهر حسة الىتى خميل كشى اال حاذ اهحيوا ي نويث اهزراػث ةيشحهر جايؼث اهزكازيق يصر * يؼهد ةحوخ اال حاذ اهحيوا ي وزارت اهزراػث ** نويث اهزراػث واهطب اهةيطرى اهدكي اهلاهرت يصر جايؼث اهيوم شؼود اهييونث اهؼرةيث اهشؼوديث حى اهح ةؤ ةاهليى اهحرةويث ويؼايالج اهدكث ةاشحخداى يوذذ اهحيوا يحؼدد اهصفاج هوطيور شواء في وجود أو ػدى وجود شجالج هدالدث شالالج ي اهدجاذ. اشحخدى ػدد 37 & 96 & 030 نحنوج ي شالالج اهي حزت اهذهةي واهي حزت اهفضي واهيطروح - ػوي اهحواهي. شجوج ةيا اج وز اهجشى هنل نحنوج ػ د ػير اهفلس 8 4 جيل واحد وحى حشاب يؼدل اهزيادت اهيوييث ةي اهفحراج ي ػير اهفلس- أشاةيغ & -8 أشةوع. أظهرت الىثبئد هبيمى: أشةوع في حجرةث هيدت أشاةيغ & - نا ج كيى اهينافئ اهورادي اهيلدر ي اهحةاي اهورادي اهيضيف هصفث وز اهجشى ػ د ػير اهفلس أػوي ػ حوم اهيلدرت في األػيار اهيحأخرت ( أشةوع( هنل ي شالهث اهي حزت اهذهةي واهي حزت اهفضي. ونا ج حوم اهليى هصفث وز اهجشى ػ د ػير اهفلس & ػ د ػير أشةوع هنل ي شالهث اهي حزت اهذهةي واهي حزت اهفضي - ػوي اهحواهي. - في الطيور الثي لهب سذالت : نا اهيدى هوليى اهحرةويث اهيح ةأ ةها هصفث وز اهجشى جى ػ د ػير اهفلس & جى ػ د ػير 4 أشاةيغ & جى ػ د ػير 8 أشاةيغ & جى ػ د ػير أشةوع هشالالج اهي حزت اهذهةي واهي حزت اهفضي واهيطروح - ػوي اهحواهي. ونا ج اهليى اهي اظرت هصفث يؼدالج اهزيادت اهيوييث هوشالالج اهدالدث ػوي اهحواهي.5 983

22 M. M. Irqi, et l جى في اهفحرت ي ػير اهفلس- 4 أشاةيغ & جى في اهفحرت ي ػير 8-4 أشاةيغ & جى في اهفحرت ي -8 أشةوع. 3- فى اآلتبء )الذكور واالىبج( الثى ليس لهب سذالت : نا اهحد األد ي واهحد األػوي هوليى اهحرةويث اهيحوكؼث واهيدى هصفث وز اهجشى ويؼدالج اهزيادت اهيوييث حأخذ فس االحجا هوطيور )اه شل( اهحي هها شجالج. 4 -شجوج شالهث اهيطروح ةصفث ػايث أػوي يدى هوليى اهحرةويث في اهطيور اهحي هها شجالج واآلةاء ( اهذنور واال اخ( اهحي هيس هها شجالج هصفاج اه يو ػ د يلار حها ةحوم اهيشجوث هشالالج اهي حزت اهفضي واهذهةي. 5 -أظهرج شالهث اهيطروح أػوي يؼايالج دكث هوحد األد ي واهحد األػوي هوليى اهحرةويث هآلةاء )اهذنور( هصفاج وز اهجشى ويؼدالج اهزيادت اهيوييث يويها شالالج اهي حزت اهفضي واهذهةي. نيا شجوج أيضا أيهاج شالالج اهيطروح واهي حزت اهفضي أػوي يؼايل دكث هوليى اهحرةويث ػ حوم اهيشجوث هوي حزت اهذهةي. ونا ج يحوشطاج يؼايالج اهدكث هوليى اهحرةويث اهيحوكؼث هنل اهشالالج واحدت حلريةا هنل اآلةاء اهذنور. 984