Trade off for fast computing - vanTeeffelenLab/ExTrack GitHub Wiki

A trade-off between number of states N, window length m, and the number of sub-steps u has to be found for reasonable computational time. When running the ExTrack fitting module with two-states on a computer with Intel® Core™ i7-9700 processor (10 000 tracks of 10 positions) with 200 iterations can be as fast as 20 seconds for a window length of m = 2 and no substep (_u_=1).

To save computational time, we recommend to initially run ExTrack with low values of u and m and then to increase u if model predictions suggest high transition rates or m for low predicted diffusion lengths. Specifically, we suggest to make the following adjustments: If localization error is negligible, for instance if there is no immobile state and all diffusion lengths d > 2 or 3 σ (localization error), window length m can be set to its minimal value of 1. Note that here the diffusion length d equals sqrt(2*D*Δt) with D the diffusion coefficient and Δ_t_ the time step. Similarly, a m = 1 should perform alright when there is one immobile state and all diffusive states have large d (> 5 σ). In such cases, multiple sub-steps can be used at little computational cost. More generally, if predicted transition rates are larger than 0.4 / Δt but localization error not negligible, we suggest increasing u to 2 for most accurate estimates. In the hardest cases of small d < 2 σ and high transition rates (> 0.4), we recommend using u = 2 and m > 8.