Multisite λ-Dynamics for Protein-DNA Binding Affinity Prediction.

Area of Science:

• Molecular Biology
• Computational Biology
• Biophysics

Background:

• Transcription factors (TFs) are crucial proteins that regulate gene expression by binding to specific DNA sequences.
• Dysregulation of TF binding is implicated in various cellular processes and disease pathways.
• Computational methods, such as λ-Dynamics, are emerging as powerful tools for predicting TF-DNA binding affinities.

Purpose of the Study:

• To evaluate the efficacy of different λ-Dynamics perturbation schemes for calculating binding free energy changes (ΔΔG).
• To assess the performance of these schemes in predicting the impact of mutations in the WRKY transcription factor's W-box binding site.

Main Methods:

• Utilized λ-Dynamics simulations to compute binding free energy changes (ΔΔG) for WRKY TF mutants.
• Compared various λ-Dynamics perturbation schemes, focusing on a single λ per base pair protocol.
• Applied the optimized protocol to additional W-box binding site mutants.

Main Results:

• The single λ per base pair protocol in λ-Dynamics exhibited the fastest convergence and highest precision.
• Calculated ΔΔG values for mutated binding sites (GATAAA, GGTCCG, GGACAA) successfully ranked relative binding affinities.
• Demonstrated the protocol's capability to accurately predict the effects of sequence variations on TF binding.

Conclusions:

• The single λ per base pair λ-Dynamics protocol is a precise and efficient method for predicting TF-DNA binding free energy changes.
• This computational approach holds significant potential for high-throughput screening and characterization of TF binding sites.