Extended Data Fig. 5: Gait and Sham.

a. Gait variables (step length, step height, and stride velocity) for SCS ON and OFF for all sessions normalized by the median value with SCS OFF across repetitions for each session. P-values corrected with Bonferroni correction (N = 2, 4 and 4 for SMA01, SMA02 and SMA03 respectively) b, Same gait variables for each session with SCS OFF. We used Kruskal Wallis ANOVA to find significance differences between groups (all p-values < 0.001). To study if the improvements saturated, we compared each week with the following week. In most gait quality variables, the improvements do not saturate (significance difference between week 3 and week 4). P-values corrected using Bonferroni correction (N = 3, 4 and 4 for SMA01, SMA02 and SMA03 respectively) c, Therapeutics effects in gait quality variables (week 1 vs week 4). All gait quality variables improved after the SCS intervention. d. Left, Examples of isometric torque traces during maximum voluntary knee extension contraction with SCS ON, SCS OFF, and with sham SCS parameters. The sham consisted of an SCS configuration chosen not to target the specific muscles involved in the joint movement tested changing the active contacts or the amplitude or frequency of the stimulation. Right, Maximum torque during knee extension for SCS ON, OFF, and sham SCS parameters. Dots represent individual repetitions, while error bars indicate mean and the 95% CI of the mean (CI). e. Maximum torque during Hip Flexion for SMA01 using the same program with different stimulation amplitudes. Dots represent individual repetitions, while error bars indicate mean and the 95% CI of the mean (CI). f. Left, Mean heel marker trajectory across gait cycles with SCS ON using same program and different frequencies. Right, gait quality variables computed under the same conditions. P-values corrected using Bonferroni correction (N = 7). g, Setup for testing exaggerated hip flexion walking on the anti-gravity treadmill (Supplementary Methods). h. Left, Examples of metatarsal marker trajectory during exaggerated hip flexion walking with SCS ON and OFF. Right, The maximum hip flexion increased with SCS ON. i, same to h, considering SCS OFF in week 1 vs week 4. Again, the maximum hip flexion increased from week 1 to week 4 even without SCS. Box plots represent the median, 25th and 75th percentile and minimum and maximum data points without outliers. All statistical significance was assessed with two-sided bootstrapping (N = 10,000) and corrected for multiple comparisons: p < 0.05 (*), p < 0.01 (**), p < 0.001(***).