Does rosetta 3.3 have different energy terms from 2.3 (fixbb design)

[Anonymous]

Dear ALL,
When I am looking the energy terms of the fixbb, I find that the rosetta 3.3 has terms as below
fa_atr 0.8
fa_rep 0.44
fa_sol 0.65
fa_intra_rep 0.004
fa_pair 0.49
fa_plane 0
fa_dun 0.56
ref 1
hbond_lr_bb 1.17
hbond_sr_bb 1.17
hbond_bb_sc 1.17
hbond_sc 1.1
p_aa_pp 0.64
dslf_ss_dst 0.5
dslf_cs_ang 2
dslf_ss_dih 5
dslf_ca_dih 5
pro_close 1.0

and
> hbond_sr_bb *= 0.5
> p_aa_pp *= 0.5
> rama = 0.2
> omega = 0.5

But 2.3 has only 11 terms.
bk_tot The total score using the design energy function. Lower is better.
fa_atr The attractive portion of the Lennard-Jones? potential. Rewards close contacts.
fa_rep Lennard-Jones repulsive term. Penalizes overlaps.
fa_sol Lazaridis-Karplus solvation model. Penalizes buried polars.
fa_dun Internal energy of sidechain rotamers as derived from Dunbrack’s statistics.
fa_intrares Intra-residue clashes.
fa_pair Statistics based paird term. Favors salt bridges.
fa_prob Probabilistic term: P(aa
hb_sc Sidechain-sidechain and sidechain-backbone hydrogen bond energy.
hb_srbb Backbone-backbone hydrogen bonds close in primary sequence.
hb_lrbb Backbone-backbone hydrogen bonds distant in primary sequence.

Also, I am confused that whether bk_tot is an energy term in rosetta2.3. It seems to me, bt_tot is the sum of packing energy. Thank you for your kindly help!

[Anonymous]

bk_tot is not a term in and of itself, it is a summarization of energies used in packing (as opposed to all energies, some of which can’t change during packing, like omega).

3.x represents a reimplementation of the 2.x scorefunction. Some terms are changed in their fine details but in broad strokes it’s the same. We don’t yet have a publication detailing the scorefunction (really, we’ve never had one) but there is one in preparation.

[Anonymous]

Thank you very much!

[Anonymous]

Dear smlewism,
I check the total score of my designed structure by rosetta 2.3. It shows
score: -586.99
But if I use rosetta3.3 score_jd2.default to score the structure, it shows -419.210. If the energy terms and parameters are the same, why the scores are different? Thank you!

[Anonymous]

They aren’t the same – “in broad strokes, they’re the same”. The scorefunctions measure the same qualities of the structure (quality of packing, hydrogen bonds, etc), but there are tweaked weights and altered implementations. The two scorefunctions aren’t meant to be compared to one another (or with physical energies, for that matter) – the scorefunctions are primarily useful for ranking within a population of models.

[Author]

Posts