Author Correction: Epigenetic stress responses induce muscle stem-cell ageing by Hoxa9 developmental signals (Nature, (2016), 540, 7633, (428-432), 10.1038/nature20603)

In this Letter, errors occurred in the following figures. In Extended Data Fig. 6e, the ‘shScr, Aged donor’ image is a duplicate of the ‘Vehicle, Aged donor’ image in Fig. 3f. The images in Extended Data Fig. 6e represent differences in engraftment levels under four experimental conditions; however, these reflect the lower end of the observed overall engraftment rate in the experiment. Figure 1 of this Amendment shows the corrected panels for Extended Data Fig. 6e, with images from the original experiment that best reflect the differences in, and the overall level of, the engraftment rates between the conditions under study (the original images from Extended Data Fig. 6e are shown for comparison). In addition, there are errors in the Source Data for Figs. 3d, 4k, Extended Data Figs. 4f–h, 7f, s, t and 9m–o, q–s due to copy-and-paste errors or due to the presentation of controls that were used for the calculation of P values or error bars shown in the figures. One value that was identified as an outlier in Extended Data Fig. 10g was not labelled as such in the original Source Data and was erronously included for graphical depiction. See Supplementary Information to this Amendment for the corrected Source Data files, and Figs. 2, 3 and 4 of this Amendment for the corrected and original panels for Figs. 3d, 4k and Extended Data Fig. 4f, g, respectively. For the calculation of the P value in Extended Data Fig. 6b, we applied a one-sided paired ratio Student’s t-test (not, as stated, a two-sided Student’s t-test). In addition, in Extended Data Fig. 7m, n, p, r, 9i, q–s and 10d, e of the original Letter, errors occurred in data scaling that affect the calculation of P values and the graphical presentation of the data. See Figs. 5, 6 and 7 of this Amendment for the corrected and incorrect panels for Extended Data Fig. 7f, m, n, p–t, 9i, m, q–s and 10d, e, g, respectively, and Supplementary Information to this Amendment for the corrected Source Data. The errors in data scaling occurred because two methods of data scaling were used throughout the study. In some experiments, data of the experimental groups were scaled to the average of the control values; in other experiments, data of the experimental groups were scaled to each of the corresponding controls of a biological repeat, set to 1 or 100. Although both methods of scaling are valid, they should not be combined within one experiment, which happened in the aforementioned figures. This has now been corrected and we include a detailed description of our scaling approach in the Supplementary Information to this Amendment. The outlined corrections do not change the conclusions of the original Letter, and we apologize for any confusion that these errors may have caused. The original Letter has not been corrected.