Time-dependent ARMA modeling of genomic sequences
2008

Modeling Non-Stationary Genomic Sequences with TD-ARMA

publication Evidence: moderate

Author Information

Author(s): Jerzy S Zielinski, Nidhal Bouaynaya, Dan Schonfeld, William O'Neill

Primary Institution: University of Arkansas at Little Rock

Hypothesis

Can a time-dependent autoregressive moving average (TD-ARMA) model effectively analyze non-stationary genomic sequences?

Conclusion

The TD-ARMA model provides a stable method for analyzing non-stationary genomic sequences, revealing that both coding and non-coding regions exhibit statistical correlations.

Supporting Evidence

  • Both coding and non-coding regions of DNA sequences are shown to be non-random.
  • The coding sequences are statistically 'whiter' than non-coding sequences.
  • The TD-ARMA model allows for a more accurate analysis of genomic sequences compared to stationary methods.

Takeaway

Scientists used a special math model to study DNA sequences and found that both important parts of DNA are not random, but the coding parts are more organized than the non-coding parts.

Methodology

The study used a time-dependent autoregressive moving average (TD-ARMA) model to analyze genomic sequences, estimating time-varying coefficients through recursive least-squares algorithms.

Limitations

The model's stability can be affected by minor perturbations in genomic data and experimental procedures.

Digital Object Identifier (DOI)

10.1186/1471-2105-9-S9-S14

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication