In this paper we predict gait deviations found in prodromal hereditary spastic paraplegia  type 4 (SPG4) based on hyperreflexia and muscle weakness in a neuro-musculoskeletal model. We found that an increased velocity feedback in the stretch reflex explains gradual gait and muscle activation changes from the prodromal to a mild-to-moderate state of SPG4. More severe gait changes, i.e. toe gait, can be predicted by a combination of hyperreflexia and muscle weakness. We conclude that neuro-musculoskeletal models are a valid tool to predict disease severity based on physiological plausible alterations in SPG4 and can be used to design and evaluate future therapeutic interventions. Our work was published in the Journal of Neuro Engineering and Rehabilitation and was presented on the International Symposium on Computer Methods in Biomechanics and Biomedical Engineering (CMBBE) and the ISPGR World Congress.