From ac02711992579042f606bf4636d68836d4eeccf0 Mon Sep 17 00:00:00 2001 From: kchu25 Date: Wed, 21 Aug 2024 17:44:33 -0400 Subject: [PATCH] update --- README.md | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 79e6039..2d92c99 100644 --- a/README.md +++ b/README.md @@ -26,13 +26,14 @@ will give # Note +### Default genomic background The information content $IC(\cdot)$ (i.e. the "letter height") of the $i$-th column $c_i$ in a position weight matrix (PWM) is $$ -IC(c_i) = \sum_{\alpha}\,f_{\alpha i}\,\log_2 \frac{f_{\alpha i}}{\beta_\alpha} +IC(c_i) = \sum_{\alpha}f_{\alpha i}\log_2 \frac{f_{\alpha i}}{\beta_\alpha} $$ -where $f_{\alpha i}$ the frequency of nucleotide $\alpha\in {A,C,G,T}$ at the $i$-th column of a PWM. The $4$-dimensional vector $\beta$ is the assumed genomic background +where $f_{\alpha i}$ the frequency of nucleotide $\alpha\in\Set{A,C,G,T}$ at the $i$-th column of a PWM. The $4$-dimensional vector $\beta$ is the assumed genomic background By default, the background $\beta$ is assumed to be flat, i.e. $\beta=(0.25, 0.25,0.25,0.25)$.