Add upcoming preprint

README.md

@@ -1,7 +1,12 @@
 # Molecular simulations for METL
 
 This repository facilitates high-throughput Rosetta runs to compute energy terms for protein variants.
+For more information, please see the [metl](https://github.com/gitter-lab/metl) repository and our manuscript:
 
+**Biophysics-based protein language models for protein engineering**.  
+Sam Gelman, Bryce Johnson, Chase Freschlin, Sameer D'Costa, Anthony Gitter<sup>+</sup>, Philip A Romero<sup>+</sup>.  
+*bioRxiv*, 2024.  
+<sup>+</sup> denotes equal contribution.
 
 ## Table of Contents
   * [Setup](#setup)
@@ -337,4 +342,4 @@ Then, add the results of one or more HTCondor runs to the database using the `da
 python code/process_run.py database --main_run_dirs output/htcondor_runs/my_condor_run --db_fn variant_database/my_database.db
 ```
 
-This database can now be used with the [metl](https://github.com/gitter-lab/metl) repository to create a processed Rosetta dataset and pretrain METL models.
+This database can now be used with the [metl](https://github.com/gitter-lab/metl) repository to create a processed Rosetta dataset and pretrain METL models.