Chimeric antigen receptor (CAR) costimulatory domains derived from native immune receptors steer the phenotypic output of therapeutic T cells. We constructed a library of CARs containing ~2300 synthetic costimulatory domains, built from combinations of 13 signaling motifs. These CARs promoted diverse human T cell fates, which were sensitive to motif combinations and configurations. Neural networks trained to decode the combinatorial grammar of CAR signaling motifs allowed extraction of key design rules. For example, non-native combinations of motifs that bind tumor necrosis factor receptor–associated factors (TRAFs) and phospholipase C gamma 1 (PLCγ1) enhanced cytotoxicity and stemness associated with effective tumor killing. Thus, libraries built from minimal building blocks of signaling, combined with machine learning, can efficiently guide engineering of receptors with desired phenotypes.

The innately heterodimeric K _2P filter permits two C-type gating mechanisms, pinching and dilation, to operate in one channel.

Summary: Cell therapies are powerful technologies in which human cells are reprogrammed for therapeutic applications such as killing cancer cells or replacing defective cells. The technologies underlying cell therapies are increasing in effectiveness and complexity, making rational engineering of cell therapies more difficult. Creating the next generation of cell therapies will require improved experimental approaches and predictive models. Artificial intelligence (AI) and machine learning (ML) methods have revolutionized several fields in biology including genome annotation, protein structure prediction, and enzyme design. In this review, we discuss the potential of combining experimental library screens and AI to build predictive models for the development of modular cell therapy technologies. Advances in DNA synthesis and high‐throughput screening techniques enable the construction and screening of libraries of modular cell therapy constructs. AI and ML models trained on this screening data can accelerate the development of cell therapies by generating predictive models, design rules, and improved designs.

The heterotrimeric SecY translocon complex is required for the cotranslational assembly of membrane proteins in bacteria and archaea. The insertion of transmembrane (TM) segments during nascent-chain passage through the translocon is generally viewed as a simple partitioning process between the water-filled translocon and membrane lipid bilayer, suggesting that partitioning is driven by the hydrophobic effect. Indeed, the apparent free energy of partitioning of unnatural aliphatic amino acids on TM segments is proportional to accessible surface area, which is a hallmark of the hydrophobic effect [Öjemalm K, et al. (2011) Proc Natl Acad Sci USA 108(31):E359–E364]. However, the apparent partitioning solvation parameter is less than one-half the value expected for simple bulk partitioning, suggesting that the water in the translocon departs from bulk behavior. To examine the state of water in a SecY translocon complex embedded in a lipid bilayer, we carried out all-atom molecular-dynamics simulations of the Pyrococcus furiosus SecYE, which was determined to be in a “primed” open state [Egea PF, Stroud RM (2010) Proc Natl Acad Sci USA 107(40):17182–17187]. Remarkably, SecYE remained in this state throughout our 450-ns simulation. Water molecules within SecY exhibited anomalous diffusion, had highly retarded rotational dynamics, and aligned their dipoles along the SecY transmembrane axis. The translocon is therefore not a simple water-filled pore, which raises the question of how anomalous water behavior affects the mechanism of translocon function and, more generally, the partitioning of hydrophobic molecules. Because large water-filled cavities are found in many membrane proteins, our findings may have broader implications.