Author Correction: Prospective de novo drug design with deep interactome learning
Kenneth Atz, Leandro Cotos, Clemens Isert, M. Håkansson, Dorota Focht, Mattis Hilleke, David F. Nippa, Michael Iff, Jann Ledergerber, Carl C. G. Schiebroek, Valentina Romeo, Jan A. Hiss, Daniel J. Merk, Petra Schneider, Bernd Kuhn, Uwe Grether, Gisbert Schneider · Nature Communications · 2025
The original version of this article contained an error in panel b of Fig. 3, in which the structure of compound 9 on the right side was incorrectly labelled as 6. The corrected version replaces the label 6 with 9. The description of panel b in the legend of Figure 3 was corrected to reflect this change and to clarify the numeric ordering of the structures. The original version stated: ‘ b Molecular structures of the five top-ranking de novo designs. The ranking criteria were PPARγ + PPARδ dual-target affinity and structural novelty (left, 1 & 6 – 9 ), or PPARγ single-target affinity and structural novelty (right, 2 , 6 , & 10 – 12 ).’ The corrected version states “ b Molecular structures of the top-ranking de novo designs. The predicted quality of the molecules decreases progressively from top to bottom. The ranking criteria were PPARγ + PPARδ dual-target affinity and structural novelty (left, 1 & 6–9 ), or PPARγ single-target affinity and structural novelty (right, 2 , & 9–12 ). Molecule 9 ranks among the top five molecules on both lists.” This has been corrected in both the PDF and HTML versions of the article.