Across the tree of life, nature is governed by fine-tuned and modulated processes at the molecular level. Proteins, especially membrane proteins, play a pivotal role in them due to their ability to undergo conformational changes; for example, cells communicate with each other through interactions between signaling molecules and receptors, the latter changing conformation upon binding of the former. The sequence-structure-function relationship is the holy grail of the study of proteins. Once deciphered, we can study this relationship under different conditions and alter the function of existing isoforms by developing drugs that target specific molecular determinants in a particular member of the family. Standard molecular dynamics (MD) simulations, while enabling essential insights at atomistic resolution, are usually too short to access rare events such as conformational transitions. Our group has acquired expertise in developing enhanced sampling techniques to study conformational ensembles. A mathematical description of the different conformational states of the system as a function of its atomic coordinates is required to obtain free energy landscapes (FELs) of conformational change; this description is referred to as a collective variable (CV). Typically, a good CV contains the slowest degrees of freedom, where the highest free energy barriers reside. We have recently demonstrated that good CVs can be inferred from family-wide state-specific co-evolving contacts between residue pairs. These CVs have been tested across various protein families last year (see activity report), and we would like to further refine the process of deriving them based on what we’ve learned. In addition to a suitable system description, broad sampling is a fundamental requirement to obtain FELs; in our group, we have developed an expert knowledge of the accelerated weight histogram (AWH) method, an efficient sampling algorithm, and we would like to use it to enhance the sampling of already familiar systems with machine-learning-based CVs. In this project, we aim to characterize the functional cycles and modulation profiles of three classes of membrane proteins: 1) G-protein coupled receptors (GPCRs), a family of allosteric proteins transducing the binding of extracellular ligands into intracellular signals; 2) sugar porters, which facilitate the transport of glucose and other sugars through the cellular membrane and thus play an essential role in maintaining sugar homeostasis; 3) voltage-gated ion channels, which regulate the electrochemical gradients across membranes and thus are involved in uncountable cellular processes.