README.md

gromacs_ff

** check http://zhenglz.blogspot.com/2017/05/fixing-bugs-in-ff14sb-port-for-gromacs.html to fix bugs on amber14sb_parmbsc1.ff

Trusted force field files for gromacs in this repo:

• https://github.com/intbio/gromacs_ff/tree/master/amber14sb_parmbsc1.ff AMBER ff with parmbsc1 corrections to simulate DNA from Gromacs web-site
• https://github.com/intbio/gromacs_ff/tree/master/amber14sb_parmbsc1_cufix.ff same ff but with CUFIX corrections
• https://github.com/intbio/gromacs_ff/tree/master/amber14sb_OL15.ff AMBER ff with OL15 refinements to simulate DNA (this is an alternative to pambsc1 as of Amber19) from Gromacs web-site
• https://github.com/intbio/gromacs_ff/tree/master/amber14sb_parmbsc1_opc_lmi.ff - this is the corresponding AMBER ff but with OPC water model and Li/Merz ions implemented by @molsim - testing is still needed
• https://github.com/intbio/gromacs_ff/tree/master/charmm36-mar2019.ff CHARMM36/36m force field from MacKerrel Lab
• https://github.com/intbio/gromacs_ff/tree/master/charmm36-mar2019_cufix.ff CHARMM36/36m ff with CUFIX corrections implemented by @molsim - testing is still needed.

General notes on available force fields

Simulating DNA

• The best source of information is usually the Amber manual. Currently, Amber 19 manual on page 38 discusses two options parmbsc1 from Orozco group and OL15 from Sponer group. The do not make a choice as to which is better, howerver, on the front page the OL15 is recommended.
• As for ions, by default the Young Cheatham 2008 ions are selected in Amber ff. Looks like it may be a good idea to replace them with CUFIX from Aksimentiev group.
• The amber14sb_parmbsc1_cufix is available in this repo.
• The amber OL15 can be downloaded from the gromacs web-site.

Simulating disordered proteins

• There is CHARMM36m (2016) ff, it is now recommended for all protein simulations, can be found at MacKerells web-page (see below). Also ports for Gromacs are there.
• In AMBER another way is to use OPC water model (see this paper https://pubs.acs.org/doi/10.1021/acs.jctc.8b01123) with 12-6 HFE Li and Merz ion parameters. We need to add a gromacs prot here ...

Simulations of proteins and protein folding

• Generally there are two options standard latest AMBER or CHARMM force fields.
• See AMBER manual for AMBER, currently the ff14SB is recommended.
• See MacKerell web-site for CHARMM http://mackerell.umaryland.edu/charmm_ff.shtml . Strictly speaking the charmm ff is updated with each new release of charmm (currently 44). However, since CHARMM36 the changes have been small, so on MacKerell's web-page they are reffered as updates to the CHARMM36 ff.

Links to external ff resources

AMBER

• http://ambermd.org/AmberModels.php
• http://ambermd.org/doc12/Amber18.pdf see section 3.
• As of 2018 - ff14SB is the recommended protein force field in Amber.
• ff14SB paper https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4821407/pdf/nihms772276.pdf
• For DNA one should use currently PARMBSC1 corrections with AMBER or OL15.
• Ion parameters are important see section 3.6. Amber currently recommends Li & Merz parameters (2015).
• CUFIX corrections by Yoo & Aksimentiev, 2018, are important for charged systems
• OL ff refinements for DNA and RNA
• OPC water model