Thank you for replay my proble is how to suite lme4 (that calculate loglike into maxium likelihood value) for have a flexible tool (wrote by professor Bates ( so an efficent tool) for build up the likelihood for posterior MCMC