Tertiary structure prediction methods MCQs With Answer

Tertiary structure prediction methods MCQs With Answer is a concise quiz set designed for M.Pharm students to strengthen their understanding of computational approaches used to predict protein tertiary structures. This set focuses on practical and theoretical aspects relevant to drug design, such as homology modeling, threading, ab initio techniques, model refinement, validation metrics and modern AI-driven methods. Each question targets conceptual clarity and application in pharmaceutical research—template selection, alignment quality, energy functions, loop and side-chain modeling, and evaluation tools used in structure-based drug discovery. Use these MCQs to assess readiness for computational projects, critical reading of literature, and integration of predicted structures into pharmacological workflows.

Q1. Which of the following best describes homology (comparative) modeling for tertiary structure prediction?

• Predicting structure solely from physical force fields and physics-based simulations
• Using experimentally determined structures of evolutionary related proteins as templates to build a model
• Assembling structures from fragments without using any template
• Predicting protein structure by docking small molecules into a sequence

Correct Answer: Using experimentally determined structures of evolutionary related proteins as templates to build a model

Q2. In homology modeling, the single most important factor for obtaining an accurate model is:

• The length of the target protein
• The resolution of the template structure
• The sequence identity between target and template
• The number of disulfide bonds in the target

Correct Answer: The sequence identity between target and template

Q3. Threading (fold recognition) differs from simple homology modeling primarily because it:

• Requires >90% sequence identity between target and template
• Attempts to fit the target sequence onto known folds even with low sequence similarity
• Uses only ab initio physics-based energy minimization
• Predicts secondary structure but not tertiary structure

Correct Answer: Attempts to fit the target sequence onto known folds even with low sequence similarity

Q4. Ab initio (de novo) structure prediction is most appropriate when:

• A high-identity template is available in the PDB
• The target sequence has many homologs with solved structures
• No reliable template exists and physics or fragment assembly must be used
• Only secondary structure information is needed

Correct Answer: No reliable template exists and physics or fragment assembly must be used

Q5. Which tool is widely used for automated homology modeling by satisfying spatial restraints derived from templates?

• Rosetta
• Modeller
• GROMACS
• BLAST

Correct Answer: Modeller

Q6. Which of the following scoring measures evaluates the overall geometry and stereochemistry of a model through backbone dihedral angles?

• Root Mean Square Deviation (RMSD)
• GDT-TS
• Ramachandran plot statistics
• pLDDT score

Correct Answer: Ramachandran plot statistics

Q7. In model assessment, GDT-TS (Global Distance Test—Total Score) is preferred over RMSD because:

• It only considers side-chain orientations
• It is less sensitive to outlier regions and better reflects global fold similarity
• It requires atomic-level experimental density maps
• It measures sequence conservation between target and template

Correct Answer: It is less sensitive to outlier regions and better reflects global fold similarity

Q8. Side-chain modeling after backbone placement is important because:

• Side chains do not affect ligand binding or active site geometry
• Accurate side-chain conformations influence packing, hydrogen bonding and docking results
• Side-chain wildcards are automatically resolved in all modeling tools
• Side chains are only needed for membrane proteins

Correct Answer: Accurate side-chain conformations influence packing, hydrogen bonding and docking results

Q9. Loop modeling in homology models is challenging because loops are:

• Highly constrained by secondary structure prediction
• Often variable in length and conformation, lacking reliable templates
• Always helical and easy to model
• Ignored in structure-based drug design

Correct Answer: Often variable in length and conformation, lacking reliable templates

Q10. Knowledge-based (statistical) potentials used in structure prediction are derived from:

• Pure quantum mechanical calculations on single atoms
• Statistical analysis of observed geometries in known protein structures
• Random number generators
• Sequence-only alignment scores

Correct Answer: Statistical analysis of observed geometries in known protein structures

Q11. Which modern AI-based method achieved a major breakthrough in tertiary structure prediction by integrating deep learning and multiple sequence information?

• GROMACS
• AlphaFold
• ClustalW
• Autodock

Correct Answer: AlphaFold

Q12. Contact prediction using co-evolutionary analysis helps tertiary prediction because:

• It directly yields exact atomic coordinates without further modeling
• Pairs of residues that co-evolve are likely to be spatially close in the folded protein
• Co-evolution only informs about membrane insertion
• It replaces the need for template selection in homology modeling

Correct Answer: Pairs of residues that co-evolve are likely to be spatially close in the folded protein

Q13. Which validation tool provides a Z-score indicating whether a model’s overall quality is consistent with experimentally determined structures?

• ProSA
• MAFFT
• HHPRED
• PSIPRED

Correct Answer: ProSA

Q14. In comparative modeling pipelines, the step most sensitive to alignment errors is:

• Energy minimization of the final model
• Template selection after model building
• Model building, because misaligned residues lead to incorrect backbone geometry
• Ramachandran plot generation

Correct Answer: Model building, because misaligned residues lead to incorrect backbone geometry

Q15. Which of the following is a typical limitation of ab initio approaches for drug discovery applications?

• They guarantee correct active site geometry for large proteins
• They are computationally expensive and less reliable for proteins >150–200 residues
• They always outperform template-based methods when templates are available
• They do not require force fields or scoring functions

Correct Answer: They are computationally expensive and less reliable for proteins >150–200 residues

Q16. In fold recognition and threading, the composite scoring function often includes which components?

• Only sequence identity score
• Compatibility of sequence with template structure, solvation/packing terms, and secondary structure agreement
• Only the number of predicted transmembrane helices
• Only molecular dynamics trajectory length

Correct Answer: Compatibility of sequence with template structure, solvation/packing terms, and secondary structure agreement

Q17. Model refinement using molecular dynamics (MD) is applied to:

• Introduce experimentally impossible conformations
• Relax steric clashes and improve local geometry while sampling realistic conformational space
• Replace the need for validation metrics
• Automatically correct poor templates without restraint

Correct Answer: Relax steric clashes and improve local geometry while sampling realistic conformational space

Q18. Which metric is specifically provided by AlphaFold models to indicate per-residue prediction confidence?

• RMSD
• pLDDT
• MolProbity score
• BLAST E-value

Correct Answer: pLDDT

Q19. For structure-based drug design, why is model validation of predicted binding site geometry critical?

• Because only backbone atoms influence ligand docking
• Because inaccurate binding-site side chains or loops can give misleading docking poses and affinity estimates
• Validation is unnecessary when the global fold looks correct
• Binding sites are always conserved and do not require assessment

Correct Answer: Because inaccurate binding-site side chains or loops can give misleading docking poses and affinity estimates

Q20. In template selection, profile–profile and HMM–HMM alignment methods are preferred because they:

• Are faster but less sensitive than pairwise BLAST
• Increase sensitivity to detect remote homologs by leveraging sequence family information
• Only work for membrane proteins
• Require experimentally solved structures of the target

Correct Answer: Increase sensitivity to detect remote homologs by leveraging sequence family information

Download