I have often observed that when ankle mobility is limited unilaterally, compensations occur during bilateral squat exercises which may result in a pelvic obliquity and in my opinion may be associated with onset or history of SI joint pain and or lumbar/ lumbosacral facet irritation. When one ankle is limited in its dorsiflexion range of motion and you descend into a depth which would would require this range several things happen. First, the unrestricted ankle is fine to allow the tibia to continue to rotate forward in front of the foot but the restricted ankle can no longer allow this forward inclination of the tibia, as a result the knee stops translating forward. In this situation experienced squatters will not allow the heel to elevate away from the floor largely because the area of pressure on the foot is an area of attention and perception which squatters will try to preserve throughout a squat. In inexperienced squatters, however, you may observe heel elevation. What does typically happen is that with the knee will no longer allowed to translate forward with further knee flexion any further knee flexion occurs with concomitant hip flexion but the hip is also forced to sit farther back due to its connection to the knee via the femur. A hip that must flex in a more posteriorly oriented position in the sagittal plane will undergo more hip flexion relative to the hip on the side of unrestricted ankle dorsiflexion simply because at a given angle of trunk inclination a hip sitting further back will make a more acute angle and require greater hip flexion. So what then results is a significant discprency in hip flexion with potential for pelvic obliquity towards the side of limited ankle dorsiflexion. The pelvic obliquity can occur because it can be difficult to maintain normal pelvic position under conditions of increased demand for end range hip flexion. Under load this pelvic obliquity may predispose one for irritation of the weight bearing structures or the structures providing stability at the SI and lumbopelvic areas. You may identify this pattern first by observing pelvic obliquity either from behind with a drop of the iliac crest or from the front as noted with a lower position of the ischial tuberosity and glute on the affected side.
The image is an example I collected with 3D motion capture and shows a large discrepancy in hip flexion likely due to decreased ankle dorsiflexion on that side. (I should now explain that those dorsiflexion values seem very large because weight bearing dorsiflexion is larger than non-weight bearing dorsiflexion values.)