Research📚💻✍🏼

Protein Folding and Association in Cellular Crowding:

The cellular environment is a complex system containing proteins, nucleic acids, lipids and metabolites, occupying nearly ~30% of its volume(~400g/L). In such a heterogeneous system, the free volume available for reactants is less than the total volume of the cell and this influences bio-molecular processes, especially the volume-dependent ones. In addition, the non-specific interactions with the environment play an active role. Such environments are clearly in contrast to in-vitro experiments typically occur in much more dilute conditions(~4-16g/L). Hence, this raises a question, "Whether the in-vivo protein stability and dynamics are different from the observed experiments?"

Despite of having a qualitative understanding of crowding effects, several questions remain poorly understood.

How does the crowding induced influence change with crowder size and shape?
What are the roles of soft long and short range interactions between crowder molecules and protein?
What is the influence of crowding on the free energy pathway for folding or association?
How mixing of various crowder species alter protein stability ?

In our group, we are developing all-atom as well as coarse-grained simulation strategies to address these questions in a systematic way.

Free energy changes for crowders of varying shapes (oligomerization states). Shown are crowder monomers with radii rD = 5 nm (green), rD = 2.5 nm (blue), dimer (cyan) and tetramer (violet). Crowder oligomers are formed by joining monomers with rD = 2.5 nm by harmonic bond and angle potentials. (Sahu et al.)

Biomolecular condensate formation by Intrinsically Disordered Proteins (IDPs):

Intrinsically Disordered Proteins(IDPs) α-Syn, TDP-43, and FUS facilitate the formation of membraneless assemblies by stabilizing dense liquid-like phases through a process called Liquid-Liquid Phase Separation(LLPS). Recently a number of IDPs have been reportedly underwent LLPS prior to formation of amyloid fibrils linked to diseases like Parkinson’s, Amyotrophic Lateral Sclerosis (ALS), Frontotemporal Dementia and Alzheimer’s.

Our lab focuses on trying to understand, how the sequence dependent cross-talk between various domains of IDPs, the network of weak multivalent interactions and environmental factors like temperature, salt concentration etc. regulate condensed phase formation by all-atom MD simulations.

S. Chakraborty and M. Biswas. "Insight into the thermo-responsive phase behavior of the P1 region of α-Synuclein using atomistic simulations" Phys. Chem. Chem. Phys., 2025,27, 5206-5214
S. Chakraborty, N. Mishra and M. Biswas. "Effect of sequence variations on the phase behavior of a functional IDP fragment" (Under Review).

Markov Modeling of Protein dynamics:

Markov State Models (MSMs) describe protein dynamics as memoryless transitions between discrete conformational states. To construct an MSM, a continuous MD trajectory is converted into discrete state-space trajectory using a suitable clustering algorithm and a Transition Probability Matrix which is estimated at a time interval(lag time) that ensures Markovianity of the dynamics. Since the transition probability matrix relies on the local density of conformations, an essential feature of the MSM is that it does not need a long continuous trajectory to predict slow dynamics of a biomolecular system, rather several relatively short trajectories can be run independently.

Our work focuses on improving several aspects of MSM model building and employ the model to study long time scale protein dynamics.

M. Biswas, B. Lickert and G. Stock, “Metadynamics Enhanced Markov Modeling of Protein Dynamics” , J. Phys. Chem. B, DOI: 10.1021/acs.jpcb.7b11800 (2018).

Data driven Langevin modeling:

In silico prediction of rare biomolecular events has been a long-standing problem, because slow (say, millisecond timescale) processes are typically out of reach of current all-atom molecular dynamics (MD) simulations. While enhanced approaches may provide an efficient sampling of the system’s free energy landscape, a dynamical description requires some coarse-grained “post-simulation” model that is capable of rebuilding the kinetics from the sampled data.

To this end, we employ the data-driven Langevin equation (dLE) approach that has been successfully shown to construct a low-dimensional dynamical model from a given MD trajectory.

N. Schaudinnus, B. Lickert, M. Biswas, G. Stock, “Global Langevin model of Multidimensional Biomolecular Dynamics”, J. Chem. Phys. 145, 184114 (2016).

Free energy landscape F(x) (in units of kBT) of AlaD, as obtained by dLE simulations (red circles) and various versions of the EVB model [Eq. (10)]. Starting with uncoupled harmonic potentials Vii(x) of the two minima [Eq. (8), dashed orange lines], the coupling V12 is varied to approximate the reference results. The cyan and brown lines show examples of too large and too small couplings and the blue line shows the optimized result. (b) Illustration of the fitting procedure of the EVB model to the dLE data (black circles). With increasing number of iterations (and increasing coupling V12), the model results change from green to red.(Schaudinnus et al.)

Google Sites

Report abuse