## LLM (Log-Linear Model)

### : inferring probabilistic conditional independency from combinatorial regulation of transcription factors

Transcription factors (TFs) regulate gene expressions by intricate combinatorial interactions. A computational modeling of the interactions is essential to understand the nature of transcriptional regulation. LLM (log-linear modeling) deals with spurious TF interactions that may lead deceptive inferences. Given discrete (0 or 1) input data of TF-DNA binding instances, this method uses a log-linear modeling that deals with probabilistic conditional independency for detecting spuriousness.

**NOTE:**

1. p-value cutoff for the test of deviance of a current RM (reduced model) from the FM (full model)

2. p-value cutoff for the test of deviance of a current RM (reduced model) from the previous RM

**Reference**

1. Lauritzen, S.L., "Graphical Models", Oxford University Press, 1996

2. Christensen, R., "Log-Linear Models and Logistic Regression", Springer-Verlag, 1997

3. Park SJ, Ichinose N, Yada T, "Probabilistic Graphical Modeling for Large-scale Combinatorial Regulation of Transcription Factors", Proc. the workshop on Knowledge, Language, and Learning in Bioinformatics (KLLBI), 72-86, 2008

4. Park SJ, Umemoto T, Saito-Adachi M, Shiratsuchi Y, Yamato M, Nakai K, "Computational Promoter Modeling Identifies the Modes of Transcriptional Regulation in Hematopoietic Stem Cells", PLoS ONE 9(4): e93853, 2014