Fig. 5: YAP regulates Oct4 transcription through modulation of burst frequency.

a Top: Two-state model of transcriptional regulation. We used this simple model to investigate where YAP interfaces with the transcription cycle to regulate downstream effectors with rate constants k_on for activation of the promoter, k_off for inactivation of the promoter, and r for transcriptional initiation. Bottom: The rate constants determine transcription burst duration, inter-burst interval, and burst amplitude. b Compound-state Hidden Markov Model-based inference of transcription burst parameters (frequency, duration, amplitude) for Oct4 in YAP KO cells, normalized to WT. p values comparing KO vs WT: frequency (Oct4: P < 0.002), amplitude (Oct4: P < 0.002) and duration (Oct4: P = 0.47). c, d Transcription burst parameters inference results for Oct4 in response to titration of steady-state YAP levels (c) or upon acute nuclear YAP export (d). High, mid and low YAP levels relate to indicated YAP concentration regimes in Fig. 4c. Dark, lit (early) and lit (late) relate to phases indicated as dark, response and adaptation in Fig. 4d, respectively. For c, p values comparing high to mid or high to low YAP levels: frequency (mid: P = 0.21; low: P < 0.002), amplitude (mid: P < 0.003; low: P = 0.007), duration (mid: P = 0.41; low: P = 0.44). For d, p values comparing lit (early) vs dark or lit (late) vs dark: frequency (lit early: P = 0.003; lit late: P = 0.29), amplitude (lit early: P = 0.1; lit late: P = 9.3 × 10⁻⁵), duration (lit early: P = 0.09; lit late: P = 0.39). Error bars in b–d reflect the standard error of the mean, as estimated from no fewer than 16 cpHMM inference replicates conducted on bootstrap samples of the experimental data. Inference analysis was performed on pooled data from N = 4 (b), N = 15 (c), and N = 12 (d) independent experiments. All p-values were calculated using 1-sided tests based on the comparison of multiple bootstrap replicates. See Methods for a detailed description of how p-values were estimated.