Protein structure predictions by enhanced conformational sampling methods

Energetic and structural insights behind calcium induced conformational transition in calmodulin

Calmodulin (CaM) is a key signaling protein that triggers several cellular and physiological processes inside the cell. Upon binding with calcium ion, CaM undergoes large scale conformational transition from a closed state to an open state that facilitates its interaction with various target protein and regulates their activity. This work explores the origin of the energetic and structural variation of the wild type and mutated CaM and explores the molecular origin for the structural differences between them. We first calculated the sequential calcium binding energy to CaM using the PDLD/S-LRA/β approach. This study  shows a very good correlation with experimental calcium binding energies. Next we calculated the calcium binding energies to the wild type CaM and several mutated CaM systems which were reported experimentally. On the structural aspect, it has been reported experimentally that certain mutation (Q41L-K75I) in calcium bound CaM leads to complete conformational transition from an open to a closed state. By using equilibrium molecular dynamics simulation, free energy calculation and contact frequency map analysis, we have shown that the formation of a cluster of long-range hydrophobic contacts, initiated by the Q41L-K75I CaM variant is the driving force behind its closing motion. This study unravels the energetics and structural aspects behind calcium ion induced conformational changes in wild type CaM and its variant.

Illuminating the "Twilight Zone": Advances in Difficult Protein Modeling

Homology modeling was long considered a method of choice in tertiary protein structure prediction. However, it used to provide models of acceptable quality only when templates with appreciable sequence identity with a target could be found. The threshold value was long assumed to be around 20-30%. Below this level, obtained sequence identity was getting dangerously close to values that can be obtained by chance, after aligning any random, unrelated sequences. In these cases, other approaches, including ab initio folding simulations or fragment assembly, were usually employed. The most recent editions of the CASP and CAMEO community-wide modeling methods assessment have brought some surprising outcomes, proving that much more clues can be inferred from protein sequence analyses than previously thought. In this chapter, we focus on recent advances in the field of difficult protein modeling, pushing the threshold deep into the "twilight zone", with particular attention devoted to improvements in applications of machine learning and model evaluation.

Computational modeling of dynein motor proteins at work

Along with various experimental methods, a combination of theoretical and computational methods is essential to explore different length-scale and time-scale processes in the biological system. The functional mechanism of a dynein, an ATP-fueled motor protein, working in a multiprotein complex, involves a wide range of length/time-scale events. It generates mechanical force from chemical energy and moves on microtubules towards the minus end direction while performing a large number of biological processes including ciliary beating, intracellular material transport, and cell division. Like in the cases of other conventional motor proteins, a combination of experimental techniques including X-crystallography, cryo-electron microscopy, and single molecular assay have provided a wealth of information about the mechanochemical cycle of a dynein. Dyneins have a large and complex structural architecture and therefore, computational modeling of different aspects of a dynein is extremely challenging. As the process of dynein movement involves varying length and timescales, it demands, like in experiments, a combination of computational methods covering such a wide range of processes for the comprehensive investigation of the mechanochemical cycle. In this review article, we will summarize how the use of state-of-the-art computational methods can provide a detailed molecular understanding of the mechanochemical cycle of the dynein. We implemented all-atom molecular dynamics simulations and hybrid quantum-mechanics/molecular-mechanics simulations to explore the ATP hydrolysis mechanisms at the primary ATPase site (AAA1) of dynein. To investigate the large-scale conformational changes we employed coarse-grained structure-based molecular dynamics simulations to capture the domain motions. Here we explored the conformational changes upon binding of ATP at AAA1, nucleotide state-dependent regulation of the mechanochemical cycle, and inter-head coordination by inter-head tension. Additionally, implementing a phenomenological theoretical model we explore the force-dependent detachment rate of a motorhead from the microtubule and the principle of multi-dynein cooperation during cargo transport.

"Dividing and Conquering" and "Caching" in Molecular Modeling

Molecular modeling is widely utilized in subjects including but not limited to physics, chemistry, biology, materials science and engineering. Impressive progress has been made in development of theories, algorithms and software packages. To divide and conquer, and to cache intermediate results have been long standing principles in development of algorithms. Not surprisingly, most important methodological advancements in more than half century of molecular modeling are various implementations of these two fundamental principles. In the mainstream classical computational molecular science, tremendous efforts have been invested on two lines of algorithm development. The first is coarse graining, which is to represent multiple basic particles in higher resolution modeling as a single larger and softer particle in lower resolution counterpart, with resulting force fields of partial transferability at the expense of some information loss. The second is enhanced sampling, which realizes "dividing and conquering" and/or "caching" in configurational space with focus either on reaction coordinates and collective variables as in metadynamics and related algorithms, or on the transition matrix and state discretization as in Markov state models. For this line of algorithms, spatial resolution is maintained but results are not transferable. Deep learning has been utilized to realize more efficient and accurate ways of "dividing and conquering" and "caching" along these two lines of algorithmic research. We proposed and demonstrated the local free energy landscape approach, a new framework for classical computational molecular science. This framework is based on a third class of algorithm that facilitates molecular modeling through partially transferable in resolution "caching" of distributions for local clusters of molecular degrees of freedom. Differences, connections and potential interactions among these three algorithmic directions are discussed, with the hope to stimulate development of more elegant, efficient and reliable formulations and algorithms for "dividing and conquering" and "caching" in complex molecular systems.