**5. Conclusions**

Predicting bound structures for IDP peptides that fold upon binding is a computational grand challenge. We have shown that possible peptide inhibitors do not necessarily bind with the same binding mode, requiring modeling approaches that allow identification of the correct binding pose. The method successfully reproduces the binding of the two inhibitors and the *p53* epitope, while showing that the two control peptides are unsuccessful binders. We further show that, by changing the intrinsic properties (e.g., helical propensity, in this case), we can identify better binders; this simplifies the design of peptide inhibitors into two distinct tasks: optimizing interface residues and optimize structural propensities. The first task requires knowing the binding mode, and the second one can be assessed by MD simulations on the free peptide, at a lower computational cost than the binding simulations. Finally, we have shown that MELD×MD is a useful tool to handle flexible binding and helps to ensure that the designed binders indeed bind and what their preferred binding mode is.

**Author Contributions:** Conceptualization, L.L. and A.P.; methodology, A.P.; software, L.L., A.P.; validation, L.L. and A.P.; formal analysis, L.L. and A.P.; investigation, L.L. and A.P.; resources, A.P.; data curation, L.L. and A.P.; writing—original draft preparation, A.P.; writing—review and editing, L.L., A.P.; visualization, L.L., A.P.; supervision, A.P.; project administration, A.P.; funding acquisition, A.P. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Data Availability Statement:** The MELD code used to run binding simulations is available to download from github: https://github.com/maccallumlab/meld .

**Acknowledgments:** This research was supported by startup funds from the Chemistry department at the University of Florida.

**Conflicts of Interest:** The authors declare no conflict of interest.
