Slow‐channel congenital myasthenic syndrome, caused by mutations in subunits of the endplate ACh receptor (AChR), results in prolonged synaptic currents and excitotoxic injury of the postsynaptic region by Ca²⁺ overloading. The Ca²⁺ overloading could be due entirely to the prolonged openings of the AChR channel or could be abetted by enhanced Ca²⁺ permeability of the mutant channels. We therefore measured the fractional Ca²⁺ current, defined as the percentage of the total ACh‐evoked current carried by Ca²⁺ ions (P_f), for AChRs harbouring the αG153S or the αV249F slow‐channel mutation, and for wild‐type human AChRs in which P_f has not yet been determined. Experiments were performed in transiently transfected GH4C1 cells and human myotubes with simultaneous recording of ACh‐evoked whole‐cell currents and fura‐2 fluorescence signals. We found that the P_f of the wild‐type human endplate AChR was unexpectedly high (P_f∼7%), but neither the αV249F nor the αG153S mutation altered P_f. Fetal human AChRs containing either the wild‐type or the mutated α subunit had a much lower P_f (2–3%). We conclude that the Ca²⁺ permeability of human endplate AChRs is higher than that reported for any other human nicotinic AChR, with the exception of α7‐containing AChRs (P_f > 10%); and that neither the αG153S nor the αV249F mutations affect the P_f of fetal or adult endplate AChRs. However, the intrinsically high Ca²⁺ permeability of human AChRs probably predisposes to development of the endplate myopathy when opening events of the AChR channel are prolonged by altered AChR‐channel kinetics.

The course of multiple sclerosis begins with a relapsing-remitting phase, which evolves into a secondarily progressive form over an extremely variable period, depending on many factors, each with a subtle influence. To date, no prognostic factors or risk score have been validated to predict disease course in single individuals. This is increasingly frustrating, since several treatments can prevent relapses and slow progression, even for a long time, although the possible adverse effects are relevant, in particular for the more effective drugs. An early prediction of disease course would allow differentiation of the treatment based on the expected aggressiveness of the disease, reserving high-impact therapies for patients at greater risk. To increase prognostic capacity, approaches based on machine learning (ML) algorithms are being attempted, given the failure of other approaches. Here we review recent studies that have used clinical data, alone or with other types of data, to derive prognostic models. Several algorithms that have been used and compared are described. Although no study has proposed a clinically usable model, knowledge is building up and in the future strong tools are likely to emerge.