FragPELE 2.2.1

Flag included to choose between srun and mpirun command to run PELE. srun is set by default, but you can also use mpirun instead.

FragPELE 2.2.0

New features:

• Parameters of the core atoms close to the fragment are modified along with growing steps, until getting the completely grown ligand.
• Parameters of theta and phi angles are also modified (for fragment and core atoms).

Both changes allow us to reduce the amount of growing steps in order to speed up the simulation time, getting slightly less accurate results in complex systems (~20%) but with an increment up to 3 folds in speed, which is fundamental for screening large libraries.

Bugs solved:

• Sequential growing failed due to messing up folder movement. Fixed on PR #47
• Some bonds had NoneType Error because its distance was missing in the bond lengths dictionary. Fixed on PR #50
• Other small bugs have been corrected

Others:

• Argparse has been re-factorized and can be call via API.
• Tests have been improved.

FragPELE 2.1.1

• Flag to not check constants.py if desired

• Fix bug when simulation failing and staying inside the working folder

FragPELE 2.1.0

New update in FragPELE! We knew that the output folders and files could be confusing sometimes and difficult to interpret. In order to solve that, we have released FragPELE2.1.0.

Major changes:

• New reorganization of FragPELE results: 1 folder for each fragment-core combination #28 #30
• New modes: only prepare and only grow. #37
• New flags added: Documents and Data symbolic links path #29

Minor changes:

• More hard-coded variables now are real variables.
• Small bugs corrected.

FragPELE 2.0.0

FragPELE 2.0

A new version of FragPELE is now available! We have done a lot of work to improve our ligand growing method and make it more suitable to be used in accordance with the real needs of the drug design market.

Major changes

• Force field parameters computation while growing. New parameters and equations applied.
• Growing onto heavy atoms: H not required anymore. More info #20
• Growing through BOND-LIKE selection. More info also in #20
• Fragments can be coupled through double and triple bonding. For more info #27
• More sampling options after growing: Explorative and personalized mode available. More info #18
• Backtracking of snapshots and connectivity: videos. Info: #31 #32

You will find all the information about how to use the new implementations in the documentation.

Minor changes

• Clash threshold now is a variable argument. More info in #10
• Distance of newly created bonds is corrected accordingly with its atom type. For more info: #21
• Some folders have changed their names: "equilibration" simulation now is the "sampling" simulation.
• The documentation includes more info, such as the analysis tools.
• An automatic analysis is performed after the simulation, generating different CSV files with all the information for each sampling simulation. Additionally, the FragPELE score for each fragment is computed in a summary TSV file.

Bugs solved

• Terminal amides now won't change from cis to trans and vice-versa. #11
• Sometimes the instruction to select a specific H could be ignored. Now, this won't happen. #17

If you feel a special interest in all the changes performed, I strongly recommend you to read the closed issues section.

Greetings,

Carles Perez

Sequential Fragments

In this new version of FragPELE we have solved some bugs presents in the very initial release and some new functions have been added.

Bugs solved:

• We erased the extra unnecessary growing step at the end of the simulation.
• The renaming system has been improved. Now allows sequential growing of fragments without any trouble.

News:

• Sequential growing completely implemented.
• Renaming. If you are not happy with the PDB atom names of your core, don't worry, you can rename them automatically with this function.
• Initial testing system. Now you can run pytest to ensure that everything is correct before starting to run FragPELE in an example.

FrAG-PELE 1.0

Break out the champagne! FrAG-PELE is already here! The new tool for in silico hit-to-lead drug design, capable of growing a fragment into a core while exploring the protein-ligand conformational space.

Some notes:

• This first version can be run with simple sh commands.
• It can be imported as Python module (use our functions, it is free).
• The results will be stored in a TSV file. Easy to read, easy to process!
• Several installation options. We strongly recommend to use Conda, but if you feel brave enough try to clone the repository in your local machine and install all the dependencies. And don't worry, soon we will be available in pip.
• If you want to know everything about FrAG-PELE, visit our documentation at https://carlesperez94.github.io/frag_pele/

We hope that you enjoy our software!

The developers Carles Perez Lopez & Dani Soler