Table 2.

Theoretical Expectations, Corresponding to Likelihoods, of Transcript Coverage with Different Models

Model/Scenario	Coverage Function f (x, l, …) =
A	$e^{\mathrm{-}l\left({\mathit{\theta }}_{\mathrm{1}}\mathrm{+}{\mathit{\theta }}_{\mathrm{2}}\right)}$
B	$\frac{1}{\left({\mathit{\theta }}_{\mathrm{1}}\mathrm{+}{\mathit{\theta }}_{\mathrm{2}}\right)}\left[{\mathit{\theta }}_{\mathrm{1}}{e}^{\mathrm{-}2l\left({\mathit{\theta }}_{\mathrm{1}}\mathrm{+}{\mathit{\theta }}_{\mathrm{2}}\right)\mathrm{+}\left({\mathit{\theta }}_{\mathrm{1}}\mathrm{+}{\mathit{\theta }}_{\mathrm{2}}\right)\left(l\mathrm{+}x\right)}\mathrm{+}{\mathit{\theta }}_{\mathrm{2}}{e}^{\mathrm{-}l\left({\mathit{\theta }}_{\mathrm{1}}\mathrm{+}{\mathit{\theta }}_{\mathrm{2}}\right)}\right]$
C	$e^{\mathrm{-}{\mathit{\theta }}_{1}l\mathrm{-}{\mathit{\theta }}_{2}x}$
D	$\frac{1}{\left({\mathit{\theta }}_{\mathrm{1}}\mathrm{+}{\mathit{\theta }}_{\mathrm{2}}\right)}\left[{\mathit{\theta }}_{\mathrm{1}}{e}^{\mathrm{-}{\mathit{\theta }}_{\mathrm{1}}\left(l\mathrm{-}x\right)}\mathrm{+}{\mathit{\theta }}_{\mathrm{2}}{e}^{\mathrm{-}{\mathit{\theta }}_{\mathrm{1}}l\mathrm{-}{\mathit{\theta }}_{\mathrm{2}}x}\right]$
E	$\frac{{\mathit{\alpha }}_1{\mathit{\alpha }}_2}{{\mathit{\theta }}_1^{\prime }\left({\mathit{\theta }}_1^{\prime }\mathrm{+}{\mathit{\theta }}_2^{\prime }\right)}\left[l\mathrm{-}\frac{1}{{\mathit{\theta }}_1^{\prime }\mathrm{+}{\mathit{\theta }}_2^{\prime }}\mathrm{-}\frac{1}{{\mathit{\theta }}_1^{\prime }}\mathrm{-}\frac{{\mathit{\theta }}_1^{\prime }{e}^{\mathrm{-}l\left({\mathit{\theta }}_1^{\prime }\mathrm{+}{\mathit{\theta }}_2^{\prime }\right)}}{{\mathit{\theta }}_2^{\prime }\left({\mathit{\theta }}_1^{\prime }\mathrm{+}{\mathit{\theta }}_2^{\prime }\right)}\mathrm{+}\frac{\left({\mathit{\theta }}_1^{\prime }\mathrm{+}{\mathit{\theta }}_2^{\prime }\right)e^{\mathrm{-}l{\mathit{\theta }}_1^{\prime }}}{{\mathit{\theta }}_1^{\prime }{\mathit{\theta }}_2^{\prime }}\right]$

Theoretical expectations (likelihood) of transcript coverage with different models (Figure 1D). x is the absolute position within the transcript (x = 0 at the 5′ end), l is the absolute transcript length (Figure 1A), and θ₁ and θ₂ are the inverse processivities of first- and second-strand syntheses, respectively. θ′₁ and θ′₂ are modified processivities for model E (see Method Details). α₁ and α₂ are the probabilities of first- and second-strand priming at a certain position, respectively. The fragmentation-related terms and parameters (d, h) are omitted for clarity.