Add better titles

README.md

@@ -80,7 +80,7 @@ For more information on the command-line arguments, run:
 protlib-designer --help
 ```
 
-### Input data : In silico deep mutational scanning data
+### Input data
 
 The input to the software is a matrix of per-mutation scores (the csv file `trastuzumab_spm.csv` in the example above). Typically, the score matrix is defined by *in silico* deep mutational scanning data, where each row corresponds to a mutation and each column corresponds to the score computed by a deep learning model. See the example data in the `example_data` directory for an example of the input data format. The structure of the input data is shown below:
 
@@ -104,7 +104,7 @@ s_{ij}^{\text{PLM}} =  -\log \left( \frac{p(x_i = a_j | w)}{p(x_i = w_i | w)} \r
 
 where $w$ is the wild-type sequence, and $p(x_i = a_j | w)$ is the probability of the mutant residue $a_j$ at position $i$ given the wild-type sequence $w$ as estimated by a Protein Language Model (PLM) or an Inverse Folding model (or any other deep learning model). For example, in [Antibody Library Design by Seeding Linear Programming with Inverse Folding and Protein Language Models](https://www.biorxiv.org/content/10.1101/2024.11.03.621763v1), we used the scores computed by the [ProtBert](https://pubmed.ncbi.nlm.nih.gov/34232869/) and [AntiFold](https://arxiv.org/abs/2405.03370) models.
 
-### Scoring functions
+### Compute Input Data using Protein Language Models
 
 We provide a set of scoring functions that can be used to compute the scores for the input data. The scoring functions are defined in the `protlib_designer/scorer` module. To use this functionality, you need to install additional dependencies: