# Frealign: Interpreting Results

A quick calculation of average scores and class sizes can be obtained using the Frealign command "frealign_calc_stats". More detailed refinement statistics are plotted at the end of each alignment parameter file. Among other tables it contains two FSC curves labeled "FSC" and "Part_FSC". The former lists the FSC as a function of resolution for the reconstruction that is masked using a spherical mask with radii "outer_radius" and "inner_radius" as specified in mparameters. The "Part_FSC" table provides an adjusted FSC curve that takes into account the amount of solvent contained in the spherical mask. This should therefore be a more accurate measure of the resolution of the 3D density of the particle, and it should be used to describe the resolution of the reconstruction. For further information, please see the paper by Sindelar & Grigorieff, 2012.

To obtain an unbiased resolution measure, it is important that the claimed resolution for the reconstruction significantly exceeds the resolution limits used during particle alignment and classification (see Multi-Reference Refinement & Classification). Provided this is the case, the FSC calculated at resolutions exceeding the resolution limits should not be significantly affected by any noise bias. If it is found in a particular case that the FSC values depend strongly on the chosen resolution limits, they may be an indication that a large portion of particles is misaligned and the resolution estimation is unreliable. In this case, it may be necessary to run additional brute-force alignment cycles, and to check that matching projections agree with the particles. Matching projections can be calculated using the Frealign command "frealign_calc_projections".