deletions | additions       diff --git a/Figure_captions_Figure_1_A__.md b/Figure_captions_Figure_1_A__.md  index 36a7772..84b1888 100644  --- a/Figure_captions_Figure_1_A__.md  +++ b/Figure_captions_Figure_1_A__.md 
...
# Figure captions  **Figure 1**: (A) Rendering PyMOL rendering  of the modeled  BglB model (cartoon) (yellow)  in complex with pNPG (spheres). The residues that were mutated in (green) showing the Cɑ of the 60 sequence positions selected for  this study are shown as red sticks. (purple).  (B) Reaction scheme of the hydrolysis of pNPG by BglB. **Figure 2**: Depiction of expression, Tm, KM, kcat, kcat/KM and conservation for each of the 100 mutants used in this study. For tm, red boxes indicate a higher Tm, and blue boxes indicate a lower Tm, as labeled in the key. For expression, a black box indicates soluble expression, and a white box indicates no expression for this mutant. For kcat, KM, and kcat/km, a diverging colormap is used, with purple values indicating lower values, and green indicating higher values.   **Figure 3**: Renderings of computational models of mutants used in this study. (A) N404C, which increased Tm by X degrees. (B) W120F, which increased melting temperature [x] degrees. (C) H178X, which … degrees. (D) XXX, which … degrees.   **Figure 4**. Receiver operating characteristic (ROC) curves use to assess machine learning predictions based on 10-fold cross validation repeated 1000 times showing our classification accuracy.References:  Fast and accurate predictions of protein stability changes upon mutations using statistical potentials and neural networks: PoPMuSiC-2.0  Predicting folding free energy changes upon single point mutations