Nyquist plots are used to understand if systems are inherently stable or unstable. This is done by plotting on the imaginary plane the poles/zeros of the transfer function of the system. If all the poles are in the negative half of the real plane then the system is stable. Otherwise, there is an element which will cause the system to go out of control or remain oscillatory (if the pole lies on the imaginary axis).
Bode plots use the transfer function in a different way in that you use a plot of the magnitude of the transfer function to determine how the system will respond to different frequency inputs. It can be viewed as a Nyquist plot evaluated on the circle |z|=1. If there's a pole that lies on this unit circle in the Nyquist plot, then the bode plot will show that there will be a frequency for which your system is unstable.