Current purposes of this repository is:
- To distribute the methods for analyzing DNA shape readout developed in Rube et al. 2017:
- To illustrate how these methods were applied to generate the data shown in figures 2,3,5 of the paper.
The files in the repository are:
-
package/computeMeanProfile.py- Computes the mean shape profile of given a list of sequences -
package/package/sampleSequences.py- Samples random sequences, computes the affinity for each sequence, and bins into affinity bins -
package/kMerLinearRegression.py- Performs linear regression on a k-mer table and outputs r, cross-validated r, regression coefficients, or the predicted values -
package/linearModelToKMer.py- Creates a k-mer table by evaluating a linear model -
package/shapeProjection.py- Performs shape projection given a mono+di binding model and a mono+di shape model -
package/shapeLibrary.py- Various functions -
scripts/fig2.sh- Generates data for Figure 2 (run make fig2) -
scripts/fig3.sh- Generates data for Figure 3 (run make fig3) -
scripts/fig5.sh- Generates data for Figure 5 (run make fig5) -
input/bindingModels- TF binding models used in Figure 2 and 5 -
input/highAffinitySequences- High-affinity sequences used for Supplemental figure 1 -
input/shapeTables- Shape tables
The shape tables were downloaded from http://rohslab.cmb.usc.edu/DNAshape/