Protein Folding Simulator: Energy Landscape & Thermodynamics

simulator advanced ~14 min
Loading simulation...
ΔG ≈ -28 kJ/mol — 99.9% folded at 310 K

A 50-residue protein with moderate hydrophobic and hydrogen bonding interactions has a folding free energy of -28 kJ/mol at 310 K (body temperature), maintaining 99.9% folded population with a melting temperature of 345 K.

Formula

ΔG = ΔH - TΔS (Gibbs free energy of folding)
f_folded = 1 / (1 + exp(ΔG / RT)) (two-state model)
Tm = ΔH / ΔS (melting temperature)

Levinthal's Paradox

A 100-residue protein has roughly 3^100 possible backbone conformations — an astronomically large number that would take longer than the age of the universe to search exhaustively. Yet proteins fold in milliseconds. Cyrus Levinthal posed this paradox in 1969 to argue that folding must follow directed pathways rather than random search. The resolution lies in the energy landscape: the funnel-shaped free energy surface ensures that the vast majority of conformational transitions are energetically downhill.

The Folding Funnel

The energy landscape theory, developed by Wolynes, Onuchic, and Dill in the 1990s, describes protein folding as progressive organization on a funnel-shaped energy surface. At the rim of the funnel, the unfolded chain has high entropy and high energy. As native contacts form, the chain descends the funnel, trading conformational entropy for stabilizing enthalpic interactions. The funnel is not smooth — roughness creates kinetic traps and folding intermediates.

Thermodynamic Stability

Protein stability is quantified by the Gibbs free energy of folding ΔG = ΔH - TΔS. The enthalpy ΔH captures the energy gained from hydrogen bonds, van der Waals contacts, and the hydrophobic effect; the entropy ΔS reflects the conformational restriction upon folding. The net stability is remarkably small — typically just 20-60 kJ/mol, equivalent to a few hydrogen bonds. This marginal stability enables the conformational flexibility required for biological function.

Temperature and Denaturation

Every protein has a melting temperature Tm where the folded and unfolded populations are equal. Above Tm, thermal energy dominates and the protein unfolds. The transition is typically highly cooperative in small proteins — resembling a molecular switch rather than a gradual unraveling. This two-state behavior is captured by the sigmoidal folded fraction curve f = 1/(1 + exp(ΔG/RT)), which transitions sharply near Tm.

FAQ

What is the protein folding problem?

The protein folding problem asks how the amino acid sequence of a protein determines its three-dimensional native structure. Despite the astronomical number of possible conformations (Levinthal's paradox), most proteins fold reliably in milliseconds to seconds. The energy landscape theory explains this through a funnel-shaped free energy surface that guides the chain toward the native state.

What is the folding energy landscape?

The energy landscape is a multidimensional surface mapping every possible protein conformation to its free energy. For foldable proteins, this landscape resembles a funnel — many high-energy unfolded states funnel down through a decreasing number of partially folded intermediates to a single low-energy native state at the bottom.

What stabilizes a folded protein?

Protein stability arises from the balance of several forces: the hydrophobic effect (burying nonpolar residues away from water), hydrogen bonds between backbone and side-chain atoms, van der Waals interactions in the tightly packed core, and electrostatic interactions including salt bridges. The net stabilization is surprisingly small — typically 20-60 kJ/mol.

How did AlphaFold change protein structure prediction?

AlphaFold2 (2020) achieved near-experimental accuracy in predicting protein structures from sequence alone, using deep learning trained on known structures and evolutionary covariance information. It effectively solved the structure prediction problem for single domains, though challenges remain for protein complexes, disordered regions, and conformational dynamics.

Sources

Embed

<iframe src="https://homo-deus.com/lab/bioinformatics/protein-folding/embed" width="100%" height="400" frameborder="0"></iframe>
View source on GitHub