I'm working on modeling evolutionary rates for 2 genes jointly in a linear relationship.  Starting with codon models I model their rates using 2 global parameters defining the line ('slope' and 'y-intercept').  I'm trying to add a new feature in which the variable "slope"  is optimized as its tangent. I want to test if this will help with convergence problems.

The error is encountered when trying to implement this.  I added the tangent function into the ReplicateConstraint command and now it will not work.  Simply removing the "Tan()" makes the script work again.  I hope this is a simple syntax error.  THANKS in advance for anyone who can help!

Here is the single offending line of code:
ReplicateConstraint ("this1.?.nonSynRate := Tan(slope) * this2.?.synRate * this3.?.nonSynRate / this4.?.synRate + yint * this5.?.synRate", Tree1, Tree1, Tree2, Tree2, Tree1);

Here is the error message I get:
Bad Binary Operator Placement:  (Tan(slope)) *? Tree1.dmel.synRate ...etc

Here is more context of the offending code:
global slope = 0;
global slope :> -3.14159;
global slope :< 3.14159;
global yint = 0;
global yint :> -1e25;
ReplicateConstraint ("this1.?.nonSynRate := Tan(slope) * this2.?.synRate * this3.?.nonSynRate / this4.?.synRate + yint * this5.?.synRate", Tree1, Tree1, Tree2, Tree2, Tree1);
Optimize (corrA_params, LF);
lnL_A0 = corrA_params[1][0];
fprintf (stdout , "corrA\t" , corrA_params[1][1] , "\t" , corrA_params[1][0] , "\tyint=" , yint,"\tslope=" , (Tan(slope)) , "\n");