Advertisement

< Back to Article

Changes in Postural Syntax Characterize Sensory Modulation and Natural Variation of C. elegans Locomotion

Fig 2

Worm locomotion sequences are complex but can be fit by a trigram model.

(A) n-gram accumulation curves for bigrams (sequences of two postures) to 5-grams (sequences of 5 postures). Over time, as more postures are observed (Total Postures), the number of unique n-grams that is observed (analogous to vocabulary size) grows sub-linearly. The grey dashed line has slope 1. Each blue line is calculated from data from 100 randomly chosen worms from the entire data set of 1262 individuals. Orange lines are averages calculated from the same data but with the sequences randomly shuffled. Black lines are averages calculated for data generated from a trigram model of the posture sequences. (B) Zipf plot of n-grams. The frequency distribution of n-grams ranked from most to least frequent. Each blue line is calculated from 400 randomly chosen worms. Black lines are averages calculated for data generated from a trigram model of the posture sequences. The red dashed line indicates the rank that divides the top 1% most frequent trigrams from the remaining 99%.

Fig 2

doi: https://doi.org/10.1371/journal.pcbi.1004322.g002