Related Experiment Videos
Parameterization studies for the SAM and HMMER methods of hidden Markov model generation
M A McClure1, C Smith, P Elton
1Department of Biological Sciences, UNLV 89129, USA.
Proceedings. International Conference on Intelligent Systems for Molecular Biology
|January 1, 1996
Summary
Generating de novo hidden Markov models (HMMs) using SAM and HMMER methods offers a flexible approach for multiple sequence alignment of distantly related viral proteins, overcoming limitations of current methods.
Area of Science:
- Computational biology
- Bioinformatics
- Genomics
Background:
- Multiple sequence alignment (MSA) is crucial for understanding protein evolution and function.
- Current MSA methods struggle with distantly related sequences, particularly viral proteins.
- Hidden Markov Models (HMMs) present a promising alternative for MSA.
Purpose of the Study:
- To evaluate the efficacy of de novo HMM generation for distantly related viral protein families.
- To compare the performance of SAM and HMMER methods in creating HMMs for diverse protein families.
- To identify optimal parameter constraints for de novo HMM construction.
Main Methods:
- Utilized SAM and HMMER software for de novo HMM generation.
- Applied HMMs to protein families including globins, kinases, aspartic acid proteases, and ribonuclease H.
- Focused on inferring appropriate parameter constraints for HMM construction.
Main Results:
- De novo HMMs demonstrate flexibility in generating multiple sequence alignments for challenging protein families.
- Both SAM and HMMER methods were employed to infer HMM parameters.
- The study details the process of parameter constraint inference for HMMs.
Conclusions:
- The hidden Markov model approach provides a flexible and effective method for multiple sequence alignment.
- This technique addresses limitations of existing methods for distantly related viral protein sequences.
- Further studies can build upon these findings for improved protein sequence analysis.