error STDEV < 0:

Dear Modellers,

I saw that someone already encoutered that error in the modeller digest .. but unfortunately there was no answer (or just to use the "output='long'" command what I did without any answer) I try to model a protein structure, using 4 different templates with about 30% of sequence identity with the target.

Modeller always send me the same error message: r_dista_644W> Two basis restraints have means too far apart:

ICSR MEAN1 MEAN2 STDEV1 STDEV2 R.VIOL1 R.VIOL2 NEW_SD1 NEW_SD2

stdevs__397E> STDEV < 0: -0.1509

Set to 1E-5.

recover____E> ERROR_STATUS >= STOP_ON_ERROR: 1 1

Could someone explain me the meaning of this and how to fix it?

thanks a lot for your help.

Muriel