Many feature transforms have been proposed for the problem of trajectory matching. These methods, which are often based on concepts from shape matching, tend to perform poorly for biological trajectories, such as cell motion, because similar biological behavior often results in dissimilar trajectory shape. Additionally, the criteria used for similarity may differ depending on the user's particular interest or the specific query behavior. We present a rank-based distance metric learning method that combines user input and a new set of biologically-motivated features for biological trajectory matching. With a small or amount of user effort in a semi-supervised framework, our method outperforms existing trajectory matching methods, even those using the same feature set.