diff --git a/R/get_jnll.R b/R/compute_nll.R
similarity index 93%
rename from R/get_jnll.R
rename to R/compute_nll.R
index cd551f4..c91bcf6 100644
--- a/R/get_jnll.R
+++ b/R/compute_nll.R
@@ -1,10 +1,11 @@
 # model function
-get_jnll <-
+compute_nll <-
 function( parlist,
           model,
           y_tj,
           family,
-          options ) {
+          options,
+          log_prior ) {
   # options[1] -> 0: full rank;  1: rank-reduced GMRF
   # options[2] -> 0: constant conditional variance;  1: constant marginal variance
 
@@ -61,7 +62,9 @@ function( parlist,
     # Full rank GMRF
     z_tj = x_tj
 
-    # Doesn't work
+    # Doesn't work ... mat2triplet not implemented
+    #V_kk = matrix(0, nrow=n_k, ncol=n_k)
+    #REPORT( V_kk )
     #V_kk = as.matrix(V_kk)
     #invV_kk = solve(V_kk)
     #Qtmp_kk = invV_kk
@@ -75,6 +78,8 @@ function( parlist,
     Q_kk = t(IminusRho_kk) %*% invV_kk %*% IminusRho_kk
 
     # Experiment
+    #IminusRho_dense = as.matrix( IminusRho_kk )
+    #V_dense = as.matrix( V_kk )
     #Q_RHS = solve(V_kk, IminusRho_kk)
     #Q_kk = t(IminusRho_kk) %*% Q_RHS
 
@@ -157,8 +162,12 @@ function( parlist,
       devresid_tj[t,j] = sign(y_tj[t,j] - mu_tj[t,j]) * pow(2 * ( (y_tj[t,j]-mu_tj[t,j])/mu_tj[t,j] - log(y_tj[t,j]/mu_tj[t,j]) ), 0.5);
     }
   }}
+
+  # Calculate priors
+  log_prior_value = log_prior( parlist )
+
   jnll = -1 * sum(loglik_tj);
-  jnll = jnll + jnll_gmrf;
+  jnll = jnll + jnll_gmrf - log_prior_value;
 
   #
   REPORT( loglik_tj )
@@ -175,6 +184,7 @@ function( parlist,
   REPORT( jnll )
   REPORT( loglik_tj )
   REPORT( jnll_gmrf )
+  REPORT( log_prior_value )
   #SIMULATE{
   #  REPORT( y_tj )
   #}
diff --git a/R/dsem.R b/R/dsem.R
index 81e7cbb..2aa5d41 100644
--- a/R/dsem.R
+++ b/R/dsem.R
@@ -264,7 +264,7 @@ function( sem,
     obj$gr_orig = obj$gr
 
     # BUild prior evaluator
-    requireNamespace(RTMB)
+    requireNamespace("RTMB")
     priors_obj = RTMB::MakeADFun( func = prior_negloglike, 
                                   parameters = list(par=obj$par), 
                                   silent = TRUE )
diff --git a/R/dsemRTMB.R b/R/dsemRTMB.R
index 1c6b94e..f99b3a2 100644
--- a/R/dsemRTMB.R
+++ b/R/dsemRTMB.R
@@ -5,7 +5,7 @@ function( sem,
           tsdata,
           family = rep("fixed",ncol(tsdata)),
           estimate_delta0 = FALSE,
-          prior_negloglike = NULL,
+          log_prior = function(p) 0,
           control = dsem_control(),
           covs = colnames(tsdata) ){
 
@@ -50,6 +50,14 @@ function( sem,
   n_j = ncol(y_tj)
   n_k = prod(dim(y_tj))
 
+  # Load data in environment for function "dBdt"
+  data4 = local({
+                  "c" <- ADoverload("c")
+                  "[<-" <- ADoverload("[<-")
+                  environment()
+  })
+  environment(log_prior) <- data4
+
   # Construct parameters
   if( is.null(control$parameters) ){
     Params = list(
@@ -110,11 +118,12 @@ function( sem,
   cmb <- function(f, ...) function(p) f(p, ...) ## Helper to make closure
   #f(parlist, model, tsdata, family)
   obj = RTMB::MakeADFun(
-          func = cmb( get_jnll,
+          func = cmb( compute_nll,
                       model = model,
                       y_tj = y_tj,
                       family = family,
-                      options = options ),
+                      options = options,
+                      log_prior = log_prior ),
           parameters = Params,
           random = Random,
           map = Map,
diff --git a/scratch/test_dsemRTMB.R b/scratch/test_dsemRTMB.R
index 435fe1b..c6c041b 100644
--- a/scratch/test_dsemRTMB.R
+++ b/scratch/test_dsemRTMB.R
@@ -1,5 +1,5 @@
 
-devtools::install_local( R'(C:\Users\James.Thorson\Desktop\Git\dsem)', force=TRUE, dep=FALSE )
+#devtools::install_local( R'(C:\Users\James.Thorson\Desktop\Git\dsem)', force=TRUE, dep=FALSE )
 
 library(dsem)
 library(RTMB)
@@ -44,11 +44,15 @@ Map$lnsigma_j = factor( rep(NA,ncol(tsdata)) )
 Params = fit0$tmb_inputs$parameters
 Params$lnsigma_j[] = log(0.1)
 
+#
+prior_negloglike = \(obj) -dnorm(obj$par[1],0,0.1,log=TRUE)
+
 # Fit model
 fit = dsem( sem=sem,
             tsdata = tsdata,
             estimate_delta0 = TRUE,
             family = rep("normal",ncol(tsdata)),
+            prior_negloglike = prior_negloglike,
             control = dsem_control( quiet=TRUE,
                                     run_model = TRUE,
                                     use_REML = TRUE,
@@ -62,20 +66,29 @@ if( FALSE ){
   covs = colnames(tsdata)
 }
 
-#
+
+###################
+# dsemRTMB
+###################
+
+# Files
 source( file.path(R'(C:\Users\James.Thorson\Desktop\Git\dsem\R)', "make_matrices.R") )
-source( file.path(R'(C:\Users\James.Thorson\Desktop\Git\dsem\R)', "get_jnll.R") )
+source( file.path(R'(C:\Users\James.Thorson\Desktop\Git\dsem\R)', "compute_nll.R") )
 source( file.path(R'(C:\Users\James.Thorson\Desktop\Git\dsem\R)', "read_model.R") )
 source( file.path(R'(C:\Users\James.Thorson\Desktop\Git\dsem\R)', "dsemRTMB.R") )
+
+# Define prior
+log_prior = function(p) dnorm( p$beta_z[1], mean=0, sd=0.1, log=TRUE)
+
 fitRTMB = dsemRTMB( sem = sem,
             tsdata = tsdata,
             estimate_delta0 = TRUE,
             family = rep("normal",ncol(tsdata)),
+            log_prior = log_prior,
             control = dsem_control( quiet = FALSE,
                                     run_model = TRUE,
                                     use_REML = TRUE,
                                     gmrf_parameterization = "projection",
-                                    trace = 1,
                                     map = Map,
                                     parameters = Params ) )
 obj = fitRTMB$obj