Loading...

Bayesian Filtering Approach to Improve Gene Regulatory Networks Inference Using Gene Expression Time Series

Fouladi, Ramouna | 2011

630 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 42386 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Fatemizadeh, Emadoddin; Arab, Shahriar
  7. Abstract:
  8. Gene regulatory modeling in different species is one of the main aims of Bioinformatics. Regarding the limitations of the data available and the perspectives which should be taken into account for modeling such networks, proposed methods up to now have not yet been successful in yielding a comprehensive model. In one of the recent researches, the Gene regulation process is considered as a nonlinear dynamic stochastic process and described by state space equations. Afterwards, in order for the unknown parameters to be estimated, Extended Kalman Filtering is used. In this thesis, first of all, Gene complexes are taken into consideration instead of genes and afterwards, Extended Kalman Filtering is used and the interactions are inferred based on the estimated parameters. Next, by considering each pair of genes together and performing the EKF algorithm on each and some changes made in the equations, the unknown parameters are estimated and interactions between genes are estimated based on the magnitude of the coefficients. Based on the simulations, by taking into consideration the gene complexes, almost 20% increase in F-Score can be seen on the S.cerevisiae dataset. In the second part, by taking into account each gene pair, the F-Score obtained shows a considerable increase in comparison with some recent approaches proposed as ARACNE, TD ARACNE BANJO and TSNI on 5 datasets. In case of Yeast (S.Cerevisiae), the Recall value has considerable increase compared to other methods. In case of the first E-Coli dataset, considerable increase is obtained in terms of PPV, Recall and F-Score compared to TSNI and BANJO. In case of IRMA Switch On dataset, PPV, Recall and F-Score have increased compared to TSNI, BANJO and ARACNE methods. In case of the IRMA Switch Off dataset also an increase around 20% compared to ARACNE, TD ARACNE, BANJO and TSNI can be seen
  9. Keywords:
  10. Extended Kalman Filter ; Bioinformatics ; Gene Regulation ; Gene Expression Data

 Digital Object List

 Bookmark

No TOC