Yes, thanks for the correction. Of course, general relativity is involved, but how it is involved ,particularly when it comes to non-biological phenomna, is completely unknown and beyond the scope of this theory.
We can test whether they evolved separately by applying analogous phenotypic traits between families and orders to different environmental niches based on similar needs.
There is a four-question survey where each practical criterion is designated by a letter (A–D) and a title in the form of a question (relating to food, predators, reproduction, and habitat).
For example, if the answer to the question “Is the common feature of this group being used differently in their habitats?” is “No,” “To be determined (TBD),” or “Not applicable (N/A),” a follow-up question is asked: “Do they respond differently in different habitats?” This may require artificially planting them in different habitats for an answer. If the answer to either question is “Yes,” we can start testing whether there are adaptive and structural convergent genes pertaining to the application of this analogous trait. If the test reveals at least one adaptive and/or structural gene, we can confidently conclude common design.
This method was inspired by a study on red and giant pandas, which concluded that their false thumbs evolved separately in response to similar needs, and a study that showed why and how they evolved separately.
Indeed, Morpho-molecular dissimilarities along with lack of fossil intermediates are good indicators of separate kinds.
For instance, the results of a recent study  confirmed the results of earlier studies, which showed that all horses are of a single created kind. Most importantly, preliminary results showed evidence that horses were, for the most part, sufficiently different from tapirs and rhinos, which share the trait of odd numbers of toes. 
NextGen Stats Confirm that All Fossil Horses Belong to the Same Created Kind (liberty.edu)
New Baraminological Methods Confirm Monobaraminic Status of the Horses (Perissodactyla: Equidae) and Preliminary Analyses of New Datasets Suggest the Possibility of Discontinuity between Horses and Various Outgroup Taxa (liberty.edu)
However, these two methods would mainly give us good reason to “suspect” all three groups are created kinds rather than only one group. Only the two-step ecology criteria I showcased above allows us to confidently conclude that they are created kinds.