When creating models to predict disease transmission, knots can appear as complications in the spread dynamics of the disease. Factors such as heterogeneous mixing patterns, super-spreader events, and local variations in public health interventions can all act as knots that need to be carefully untangled for accurate model predictions.