Skip to main content
Figure 4 | BMC Bioinformatics

Figure 4

From: OHMM: a Hidden Markov Model accurately predicting the occupancy of a transcription factor with a self-overlapping binding motif

Figure 4

Trained z is inversely proportional to the length of the training promoter. HMMs were trained on TSS-n promoters keeping the initial motif profile fixed. The transition probability to the motif (z) is inversely proportional to the training promoters' length in the range between 500–3000 bp and hence z* is constant around 0.9. This quantity drops slightly between 500 to 200 bp and then substantially after 200 bp due to the lack of κBsites in the shorter training promoters.

Back to article page