MATLAB practice in Computational Chemistry

Browse the rooms before signing in. All rooms are free with an account. Practice does not issue certificates.

• Keep Feasible Candidates

  A descriptor must fall between its lower and upper bound. Return the inclusive pass mask.

  constraints. Free with an account.

• Normalize State Populations

  Raw state weights need to become probabilities that sum to one.

  normalization. Free with an account.

• Screen a Charge Pair

  A dielectric medium weakens a point-charge interaction. Return q1*q2/(epsilon*r).

  electrostatics. Free with an account.

• Feel the Thermal Scale

  A lab needs kBT at a new temperature before comparing molecular energies.

  thermal-scale. Free with an account.

• Solvate the Ion

  Estimate continuum ion stabilization with a simplified Born expression.

  born-model. Free with an account.

• Average the Ensemble

  State energies and probabilities describe an ensemble. Return the probability-weighted energy.

  ensemble-average. Free with an account.

• Score a Tradeoff

  Two normalized objectives have declared weights. Return the weighted utility.

  utility. Free with an account.

• Integrate the Coupling Path

  A free-energy derivative was measured across lambda. Integrate it by trapezoids.

  quadrature. Free with an account.

• Accept the Monte Carlo Move

  Compare a supplied random variate with an uphill Boltzmann probability.

  metropolis. Free with an account.

• Audit Surrogate Residuals

  Observed and predicted endpoints need RMSE and worst absolute error.

  model-validation. Free with an account.

• Count Effective Samples

  A correlated trajectory contains less information than its frame count. Return n/tau capped at n.

  effective-sample-size. Free with an account.

• Find the Tradeoff Front

  For two maximized objectives, flag candidates not dominated by any other candidate.

  pareto-front. Free with an account.

• Build a Free-Energy Profile

  Convert positive histogram weights to a shifted -log probability profile.

  pmf. Free with an account.

• Close the Thermodynamic Cycle

  Directed free-energy legs should sum to zero. Return absolute closure error.

  cycle-closure. Free with an account.

• Measure Applicability Distance

  A candidate should be compared with the nearest training descriptor row.

  applicability-domain. Free with an account.

• Challenge the Constraint

  Use mean plus two standard deviations against an upper limit.

  robust-feasibility. Free with an account.

• Challenge the Surrogate

  Compare validation and training RMSE. Flag a generalization gap above twofold.

  generalization. Free with an account.

• Diagnose Weight Collapse

  Exponential reweighting needs its effective sample size. Stabilize the weights before calculating it.

  reweighting. Free with an account.

• Publish the Bounded Status

  Prioritize invalid input, sampling failure, model failure, and experimental readiness.

  evidence-dossier. Free with an account.

• Select a Diverse Batch

  After the lead candidate, choose the descriptor point farthest from it.

  experimental-design. Free with an account.