Principal Component Analysis allows a quantitative description of shape variability with a restricted number of parameters (or modes) which can be used to quantify the difference between two shapes through the computation of a modal distance. A statistical test can then be applied to this set of measurements in order to detect a statistically significant difference between two groups. We have applied this methodology to highlight evidence of genetic encoding of the shape of neuroanatomical structures. To investigate genetic constraint, we studied if shapes were more similar within 10 pairs of monozygotic twins than within interpairs and compared the results with those obtained from 10 pairs of dizygotic twins. The statistical analysis was performed using a Mantel permutation test. We show, using simulations, that this statistical test applied on modal distances can detect a possible genetic encoding. When applied to real data, this study highlighted genetic constraints on the shape of the central sulcus. We found from 10 pairs of monozygotic twins that the intrapair modal distance of the central sulcus was significantly smaller than the interpair modal distance, for both the left central sulcus (Z = -2.66; P < 0.005) and the right central sulcus (Z = -2.26; P < 0.05). Genetic constraints on the definition of the central sulcus shape were confirmed by applying the same experiment to 10 pairs of normal young individuals (Z = -1.39; Z = -0.63, i.e., values not significant at the P < 0.05 level) and 10 pairs of dizygotic twins (Z = 0.47; Z = 0.03, i.e., values not significant at the P < 0.05 level).
Copyright 2000 Academic Press.