The programs support mixed models with multiple-correlated effects, multiple animal models and dominance.

**Most** of the programs now support being called with the parameter file as an argument, e.g.

blupf90 renf90.par

This avoids typing the name of the file in interactive or doing the trick.

echo renf90.par | blupf90

Download BLUPF90 family manual. Note that some programs shown below are not in the manual (e.g. `inbupgf90`

)

How to cite the BLUPF90 manual:
Misztal I., Tsuruta S., Lourenco D.A.L., Aguilar I., Legarra A., and Vitezica Z. 2014. Manual for BLUPF90 family of programs.

## Programs

**Available for research (free with restriction: # genotyped animals = 25,000)**

- BLUPF90+ - a combined program of blupf90, remlf90, and airemlf90
- GIBBSF90+ - a combined program of gibbs1f90, gibbs2f90, gibbs3f90, thrgibbs1f90, and thrgibbs3f90
- POSTGIBBSF90 - statistics and graphics for post-Gibbs analysis (S. Tsuruta)
- RENUMF90 - a renumbering program that also can check pedigrees and assign unknown parent groups; supports large data sets
- PREGSF90 – genomic preprocessor that combines genomic and pedigree relationships (I. Aguilar)
- POSTGSF90 – genomic postprocessor that extracts SNP solutions after genomic evaluations (single step, GBLUP) (I. Aguilar)
- PREDICTF90 - a program to calculate adjusted y, y_hat, and residuals (I. Aguilar)
- PREDF90 - a program to predict direct genomic value (DGV) for animals based on genotypes and SNP solution
- QCF90 - a quality-control tool on genotypes and pedigree information (Y. Masuda)
- INBUPGF90 - a program to calculate inbreeding coefficients with incomplete pedigree (I. Aguilar)
- SEEKPARENTF90 - a program to verify paternity and parent discovery using SNP markers (I. Aguilar)
- VALIDATIONF90 - a program to perform validation of predictions (coming soon!)
- RRMEBVF90 - postprocessor for random regression models

**Available by request**

- QXPAK - joint analysis of QTL and polygenic effects (M. Perez-Enciso) QxPak web page
- MRF90 - Method R program suitable for a large data set; contact T. Druet.
- COXF90 – Bayesian Cox model - contact J. P. Sanchez (JuanPablo.Sanchez@irta.cat)
- BLUPF90HYP – BLUPF90 with hypothesis testing (F and Chi2 tests) - contact J. P. Sanchez as above

**Available only under research agreement**

- BLUP90IOD2 - BLUP by iteration on data with PCG for very large models (S. Tsuruta)
- BLUP90IOD3 (new BLUP90IOD) - a combined program of BLUP90IOD2, BLUP90IOD2RR, BLUP90IOD2HR, and BLUP90MBE2 with new features
- CBLUP90IOD2 - BLUP by iteration on data for threshold-linear models (working for one threshold and several linear traits)
- CBLUP90MBE - BLUP by iteration on data for threshold-linear models (working for one threshold and several linear traits) with support for very large models for multi-breed evaluations
- ACCF90 - approximation of accuracies for breeding values
- ACCF90GS - approximation of accuracies for genomic breeding values based on diagonals of
**G** - ACCF90GS2 - new approximation of accuracies for genomic breeding values based on block sparse inversion
- BLUP90ADJ - BLUP data preadjustment tool
- DEPROOFSF90 - computation of deregressed proof (DRP)
- DYDF90 - computation of daughter yield deviation (DYD) known as progeny trait deviation

**No longer updated (as of May 2022)**

- BLUPF90 - BLUP in memory
- REMLF90 - accelerated EM REML
- AIREMLF90 - Average Information REML with several options including EM-REML and heterogeneous residual variances (S. Tsuruta)
- GIBBSF90 - simple block implementation of Gibbs sampling - no genomic
- GIBBS1F90 - as above but faster for creating mixed model equations only once
- GIBBS2F90 - as above but with joint sampling of correlated effects
- GIBBS3F90 - as above with support for heterogeneous residual variances
- THRGIBBSF90 - Gibbs sampling for any combination of categorical and linear traits (D. Lee) - no genomic
- THRGIBBS1F90 - as above but simplified with several options (S. Tsuruta)
- THRGIBBS3F90 - as above with heterogeneous residual variances for linear traits
- CBLUP90 - solutions for bivariate linear-threshold models (works only for 2 trait model: threshold and linear)
- CBLUP90THR - as above but with thresholds computed and many linear traits (B. Auvray)
- CBLUP90REML - as above but with quasi REML (B. Auvray) - no genomic
- GIBBS2CEN - support censored data (J. Arango) - no genomic
- GIBBS2RECUR - support recursive models (J. P. Sanchez) - no genomic
- BLUP90IOD2RR & BLUP90IOD2HR - BLUP90IOD2 with heterogeneous residual variances (like in GIBBS3F90 & AIREMLF90)
- BLUP90MBE - BLUP by iteration on data with support for very large models for multi-breed evaluations (A. Legarra)

Other programming contributions were made by Miguel Perez-Enciso `(user_file)`

and François Guillaume (Jenkins hashing functions).