In all literature describing aberrations which I met, I usually saw their relation to the Zernike polynomials? Is there are other methods describing aberrations with the same efficiency?
Zernike polynomials represents optimum balanced aberrations in the circular pupil and the expansion coefficients do not show crosslink, as it is orthonormal set of functions (for any other pupil geometries a set of useful polynomials can be obtained). Further, they provide useful optical information of the wavefront. several of these polynomials are well related with optimum balanced third order classical aberrations having minimum wavefront variance, and thus maximizing Strehl ratio.
You can use any other orthonormal polynomial set but their optical usefulness has to be demonstrated. Mathematically, you can expand the wavefront aberration in any of these sets and get an accurate approximation. However, it is true that Zernike polynomials are widely used in the optical community as a standard specification and characterization.