Real-world Evidence to Advance Community Health

Anthony Gitter, PhD – Genetic and Epigenetic Mechanisms (GEM)

NIH NIGMS R01GM135631

• Given sufficient training data, simple machine learning models can be very effective in finding active molecules for new drug targets.
• Using prior screening data from SMSF, we trained a random forest classifier to predict compounds that inhibit the interaction between PriA helicase and the single-stranding binding protein. The model was tested prospectively on two large commercial chemical libraries.
• First, we virtually screened Aldrich Market Select chemical library (7 million molecules). Among the top compounds scored by the model, we purchased and tested 701 diverse candidates. Of these, 48% were active. In comparison, the hit rate in HTS was only 0.13%.
• Next, looking at performance on an ultra-large library, we applied the model to EnamineREAL (1.08 billion molecules). We purchased 68 diverse candidates prioritized by the model and observed activity in 31 compounds (46%), including one with an IC50 of 1.3 uM.

Alnammi M, Liu S, Ericksen SS, Ananiev GE, Voter AF, Guo S, Keck JL, Hoffmann FM, Wildman SA, Gitter A. ChemRxiv, 2021. doi:10.26434/chemrxiv-2021-fg8z9

Liu S, Alnammi M, Ericksen SS, Voter AF, Ananiev GE, Keck JL, Hoffmann FM, Wildman SA, Gitter A. J Chem Inf Model. 2019:59(1):282-293. doi: 10.1021/acs.jcim.8b00363.

Michael A. Newton, PhD – Genetic and Epigenetic Mechanisms (GEM)

NIH NIGMS R01GM135631

• What is the minimal compound testing data required to learn the broader ligand selectivity of some new drug target?
• Given a well-studied target class, we examined strategies for choosing small, information-rich probe sets or “informer sets” that, upon testing, provide key data to enable inference of a ligand selectivity profile for some new member of the target class.
• As a proof-of-concept, we developed and tested selection strategies that enable effective predictions of kinase inhibitor selectivity profiles on new kinase targets using probe sets of only 16 molecules.
• Excellent performance was observed both in retrospective tests on hundreds of human protein kinases and in prospective tests on 2 novel protein kinases from microbial pathogens.

Zhang H, Ericksen SS, Lee C-p, Ananiev GE, Wlodarchak N, Yu P, Mitchell JC, Gitter A, Wright SJ, Hoffmann FM, Wildman SA, Newton MA. PLOS Comp Biol. 2019. doi: 10.1371/journal.pcbi.1006813

Yu P, Ericksen SS, Gitter A, Newton MA. Biometrics 2022;1-3. doi: 10.1111/biom.13637

John Denu, PhD – Genetic and Epigenetic Mechanisms (GEM)

NIH R01 GM059785

• Overexpressed UHRF1 has been observed in numerous cancers
• HTS of 100K compounds for inhibitors of UHRF1
• AlphaScreen in 1536 well plates
• Three lead compounds identified

Wallace H Liu, Robert E Miner 3rd,  Brittany N Albaugh, Gene E Ananiev, Scott A Wildman, John M Denu. Biochemistry 2022 Mar 1;61(5):354-366. doi: 10.1021/acs.biochem.1c00698. Epub 2022 Feb 10

id Andes, MD and Tim Bugni, PhD – Infectious Diseases

U19 AI109673 and U19 AI142720

• A novel anti fungal agent Turbinmicin discovery from marine microbiome using MDR drug discovery platform.
• Turbinmicin displays potent in vitro and in vivo efficacy against multiple MDR fungal pathogens.
• MOA hypothesis is that turbinmicin impairs vesicle-mediated trafficking by inhibiting Sec14p.

Zhang F, Zhao M, Braun DR, Ericksen SS, Piotrowski JS, Nelson J, Peng J, Ananiev GE, Chanana S, Barns K, Fossen J, Sanchez H, Chevrette MG, Guzei IA, Zhao C, Guo L, Tang W, Currie CR, Rajski SR, Audhya A, Andes DR, Bugni TS. Science. 2020 Nov 20;370(6519):974-978. doi: 10.1126/science.abd6919.

hristian Capitini, MD – Developmental Therapeutics Program (DT)

• Management for high-risk neuroblastoma (NBL) has included autologous hematopoietic stem cell transplant (HSCT) and anti-GD2 immunotherapy, but survival remains around 50%.
• Allogeneic HSCT alone was insufficient to control NXS2 tumor growth, but the addition of hu14.18-IL2 controlled tumor growth and improved survival.

Bates PD, Rakhmilevich AL, Cho MM, Bouchlaka MN, Rao SL, Hales JM, Orentas RJ, Fry TJ, Gilles SD, Sondel PM, Capitini CM. Front Immunol. 2021 Aug 19;12:668307. doi: 10.3389/fimmu.2021.668307. eCollection 2021.