Primero hay que convertir la base de datos a formato Merlin. Partimos del ultimo merge realizado en Imputando 1000 Genome con plink.

La conversion se hace con fcgene,

$ fcgene --ped admurcia_merged5_chr22_noesv.ped --map admurcia_merged5_chr22_noesv.map --oformat mach --out admurcia_2mach2_chr22

||============================================================================||
*-> Copyright:  GNU General Public License
*-> Program Developer:  
         Nab Raj Roshyara        
         email: roshyara@yahoo.com       
         Universitaet Leipzig            
         Leipzig Research Center for Civilization Diseases(LIFE)
         Institute for Medical Informatics, Statistics and Epidemiology (IMISE)
         Group: Genetical Statistics and Systems Biology 
         Group Leader: Dr. rer. nat. Markus Scholz
||============================================================================||

*->Used command options: 

         --ped admurcia_merged5_chr22_noesv.ped
         --map admurcia_merged5_chr22_noesv.map
         --oformat mach
         --out admurcia_2mach2_chr22

*->Reading map file "admurcia_merged5_chr22_noesv.map":

*-> Reading  ped file: "admurcia_merged5_chr22_noesv.ped":
**->Total number of successfully read individuals: 1467
                   **->Total number of successfully read: 494091


*->Individual summary:
        **->Total female:                                 577
        **->Total male:                                   511
        **->Total undefined individual:                   379

        **->Total cases:                                  319
        **->Total controls:                               769
        **->Total undefined individuals:                  379

*->Writing files in mach format: 
         **->MACH dat file has been written out and saved as  "admurcia_2mach2_chr22.dat".
        **->MACH ped file has been written out and saved as "admurcia_2mach2_chr22.ped".
         **->One extra "snpinfo.txt" file has been written out and saved as "admurcia_2mach2_chr22_snpinfo.txt". 
        **->One extra pedinfo.txt file has also been written out and saved as "admurcia_2mach2_chr22_pedinfo.txt".

*->The analysis has ended at Thu May 23 16:32:47 2013

*->Total time taken for the analysis is: 0 hours, 15 minutes and 51.00 seconds.

Ahora hay que preparar los archivos para ejecutar minimac,

$ mach1 -d admurcia_2mach2_chr22.dat -p admurcia_2mach2_chr22.ped --rounds 20 --states 200 --phase --interim 5 --sample 5 --compact

Los resultados quedan en mach1.out.gz

Voy a probar ahora a ejecutar minimac en varios procesadores

$ $ minimac-omp --cpus 20 --vcfReference --refHaps /nas/data/GNlib/1000Genome/2mach/chr22.phase1_release_v3.20101123.snps_indels_svs.genotypes.refpanel.EUR.vcf.gz --haps mach1.out.gz --snps admurcia_2mach2_chr22_target.snps --rounds 5 --states 200 --prefix admurcia_2mach2_chr22_imputed

Esto produce 5 archivos

admurcia_2mach2_chr22_imputed.dose  
admurcia_2mach2_chr22_imputed.erate  
admurcia_2mach2_chr22_imputed.info  
admurcia_2mach2_chr22_imputed.info.draft  
admurcia_2mach2_chr22_imputed.rec