Extensive protein and DNA backbone sampling improves structure-based specificity prediction for C2H2 zinc fingers

Area of Science:

• Molecular Biology
• Structural Biology
• Bioinformatics

Background:

• Sequence-specific DNA recognition by gene regulatory proteins is essential for cellular function.
• Predicting DNA binding preferences from protein sequences is vital for reconstructing regulatory interactions.
• Structural modeling offers a pathway to predict DNA binding profiles.

Purpose of the Study:

• To develop and present a novel molecular modeling protocol for predicting protein-DNA binding preferences.
• To infer DNA binding preferences by exploring sequence space coupled with conformational sampling.
• To validate the protocol's efficacy using C2H2 zinc finger transcription factors.

Main Methods:

• Utilized conformational sampling techniques from de novo protein structure prediction.
• Generated diverse structural models of protein-DNA complexes from small fragments.
• Coupled extensive conformational sampling with sequence space exploration.

Main Results:

• Successfully predicted DNA binding profiles for eleven C2H2 zinc finger transcription factors.
• Achieved good agreement between predicted profiles and experimental binding data.
• Provided structural insights into observed DNA binding preferences.

Conclusions:

• The novel molecular modeling protocol accurately predicts protein-DNA binding preferences.
• The method aids in understanding gene regulatory protein interactions.
• Structural modeling combined with sequence exploration is a powerful approach for predicting binding profiles.