You are here

Running Rosetta CM Hybridize step errors with: Scorefunction not set up for nonideal/Cartesian scoring

8 posts / 0 new
Last post
Running Rosetta CM Hybridize step errors with: Scorefunction not set up for nonideal/Cartesian scoring
#1

I am running the example protocol (RosettaCM multi-template comparative modeling) supplied by the Meiler lab (I think?). Only I am using a different starting protein that the one provided. I have successfull made it to the hybridize step:

rosetta_scripts.static.linuxgccrelease -database $ROSETTA_DATABASE -in:file:fasta /inputs/2yhd_model.fasta -parser:protocol /inputs/2yhd_model_threaded.xml -relax:minimize_bond_angles -relax:minimize_bond_lengths -relax:jump_move true -default_max_cycles 200 -relax:min_type lbfgs_armijo_nonmonotone -relax:jump_move true -score:weights stage3.wts -use_bicubic_interpolation -hybridize:stage1_probability 1.0 -nstruct 3

but am running into an execution error.

Two questions:

(1) Should the nstruct be increased? Most Rosetta protocols use (in a real run) use nstruct >= 1000 or so. Then if I do that, the following relax step would have nstruct >= 1000'ish also? What is the best practices for the number of structures to generate on this hybridize step?

(2) This question is more important. When running the above command, rosetta scripts errors with the error:

protocols.hybridization.CartesianHybridize: RUNNING FOR 4 MACROCYCLES
protocols.hybridization.CartesianHybridize:  total_res=249   prot_res=249
protocols.hybridization.CartesianHybridize: CYCLE 1
protocols.hybridization.CartesianHybridize:   setting bonded weight = 0
protocols.hybridization.CartesianHybridize:   setting bonded angle weight = 0
protocols.hybridization.CartesianHybridize:   setting bonded length weight = 0
protocols.hybridization.CartesianHybridize:   setting bonded torsion weight = 0
protocols.hybridization.CartesianHybridize:   setting cst    weight = 0.5
protocols.hybridization.CartesianHybridize:   setting vdw    weight = 0.02

ERROR: Scorefunction not set up for nonideal/Cartesian scoring
ERROR:: Exit from: src/core/optimization/CartesianMinimizer.cc line: 64
[0x312d151]
[0x491c7ff]
[0x2f7eff5]
[0x1284f6c]
[0x1169361]
[0x2870f98]
[0x287355b]
[0x2873e7e]
[0x2a567af]
[0x2a57fba]
[0x2b2297a]
[0x41243e]
[0x4c2a8d4]
[0x95250d]
protocols.rosetta_scripts.ParsedProtocol: Exception while processing procotol:
protocols.jd2.JobDistributor: 

 

I found another topc here that had the same error, and the recommeded answer was to include cart_bonded in the score function, so updated my rosetta script XML to:

<dock_design>
    <TASKOPERATIONS>
    </TASKOPERATIONS>
    <SCOREFXNS>
        <stage1 weights="stage1.wts" symmetric=0>
            <Reweight scoretype=cart_bonded weight=0.5/>
            <Reweight scoretype=atom_pair_constraint weight=1/>
        </stage1>
        <stage2 weights="stage2.wts" symmetric=0>
            <Reweight scoretype=cart_bonded weight=0.5/>
            <Reweight scoretype=atom_pair_constraint weight=0.5/>
        </stage2>
        <fullatom weights="stage3.wts" symmetric=0>
            <Reweight scoretype=cart_bonded weight=0.5/>
            <Reweight scoretype=atom_pair_constraint weight=0.5/>
        </fullatom>
    </SCOREFXNS>
    <FILTERS>
    </FILTERS>
    <MOVERS>
        <Hybridize name=hybridize stage1_scorefxn=stage1 stage2_scorefxn=stage2 fa_scorefxn=fullatom batch=1 stage1_increase_cycles=1.0 stage2_increase_cycles=1.0 linmin_only=1>
            <Fragments 3mers="/inputs/2yhd_model.200.3mers" 9mers="/inputs/2yhd_model.200.3mers"/>
            <Template pdb="/inputs/2yhd_model_threaded.pdb" cst_file="AUTO" weight=1.000 />
        </Hybridize>
    </MOVERS>
    <APPLY_TO_POSE>
    </APPLY_TO_POSE>
    <PROTOCOLS>
        <Add mover=hybridize/>
    </PROTOCOLS>
</dock_design>

 

Running the same command as above now yields a different error:

...
protocols.hybridization.CartesianHybridize: RUNNING FOR 4 MACROCYCLES
protocols.hybridization.CartesianHybridize:  total_res=249   prot_res=249
protocols.hybridization.CartesianHybridize: CYCLE 1
protocols.hybridization.CartesianHybridize:   setting bonded weight = 0
protocols.hybridization.CartesianHybridize:   setting bonded angle weight = 0
protocols.hybridization.CartesianHybridize:   setting bonded length weight = 0
protocols.hybridization.CartesianHybridize:   setting bonded torsion weight = 0
protocols.hybridization.CartesianHybridize:   setting cst    weight = 0.5
protocols.hybridization.CartesianHybridize:   setting vdw    weight = 0.02
core.scoring.CartesianBondedEnergy: Creating new peptide-bonded energy container (249)

ERROR: cartfrag_overlap_>=1 && cartfrag_overlap_<=len/2
ERROR:: Exit from: src/protocols/hybridization/CartesianHybridize.cc line: 341
[0x312d151]
[0x491c7ff]
[0x128360b]
[0x1284d74]
[0x1169361]
[0x2870f98]
[0x287355b]
[0x2873e7e]
[0x2a567af]
[0x2a57fba]
[0x2b2297a]
[0x41243e]
[0x4c2a8d4]
[0x95250d]
protocols.rosetta_scripts.ParsedProtocol: Exception while processing procotol:
protocols.jd2.JobDistributor: 

 

What additional information should I post to help diagnose this issue? The weights are attacted (and relabled as .txt so I can upload).

 

AttachmentSize
stage1.wts_.txt182 bytes
stage2.wts_.txt257 bytes
stage3.wts_.txt536 bytes
Post Situation: 
Wed, 2017-03-29 13:19
brspurri

I have a same problem (2019 rosetta).

Sat, 2019-12-07 13:12
dnamkr

I have a same problem (2019 rosetta).

Sat, 2019-12-07 13:12
dnamkr

Hi,

I'm not sure if this will work for your application, but you might try adding -score:set_weights pro_close 0 cart_bonded 0.5

I run things from a command line, not a script, so I'm not sure how these flags would translate into your script.

I've been trying to do Cartesian and Dualspace Relax, and Frank DiMaio helped me get them running with this extra flag!

Here's the full Dualspace Relax command that worked for me:

$HOME/ROSETTA_2019_22/main/source/bin/relax.default.linuxgccrelease -in:file:s 16pk.cif -score:set_weights pro_close 0 cart_bonded 0.5 -default_max_cycles 200 -dualspace -nstruct 5 -out:suffix _relaxed -out:mmCIF -out:path:mmCIF Output_CIFs/16pk -out:path:score Output_Scores/16pk > Output_Logs/16pk/16pk_Relax.log

Hope it helps,

AJ Vincelli

 

Thu, 2019-12-12 07:27
AJVincelli

May I ask what is the score function you're using? Is it cart_bonded or atom_pair_constraint?

Thu, 2020-03-05 10:07
Morrian_Lynn

Hi Morrian, I'm sorry that I just saw your comment! (I'm going to have to check my notification settings...) In case this info is helpful despite being so late: I'm just using the default score function, which is ref2015 for the version of Rosetta that I'm using. The "cart_bonded 0.5" flag turns on this score term I believe, and the "pro_close 0" turns off this score term (a weight of 0). Let me know if you want more info!

Sun, 2021-07-18 16:50
AJVincelli

Hi Morrian, I'm sorry that I just saw your comment! (I'm going to have to check my notification settings...) In case this info is helpful despite being so late: I'm just using the default score function, which is ref2015 for the version of Rosetta that I'm using. The "cart_bonded 0.5" flag turns on this score term I believe, and the "pro_close 0" turns off this score term (a weight of 0). Let me know if you want more info!

Sun, 2021-07-18 16:48
AJVincelli

There is a dedicated cartesian scorefunction version of the ref2015, called ref2015_cart in the database, which already has the following difference:

| name         |   pro_close |   cart_bonded |
|:-------------|------------:|--------------:|
| ref2015      |        1.25 |           0   |
| ref2015_cart |        0    |           0.5 |

Also note that 0.8 & 0.44 for the Lenard-Jones as seen in the stage3.wts is Score12 and Talaris2013 as opposed to 1.0 & 0.55 for Talaris2014 and ref2015.
But its actually Score12 derived, as seen for solvatation and the residue specific ∆∆G fudge factors (REF). Unfortunately my code to compare scorefxns does not allow loading from outside of of the database, so I cannot load the stage3 but here are the regular scorefunctions:

  fa_atr fa_rep fa_sol fa_intra_rep pro_close fa_pair hbond_sr_bb hbond_lr_bb hbond_bb_sc hbond_sc dslf_ss_dst dslf_cs_ang dslf_ss_dih dslf_ca_dih rama omega fa_dun p_aa_pp ref_ALA ref_ARG ref_ASN ref_ASP ref_CYS ref_GLN ref_GLU ref_GLY ref_HIS ref_ILE ref_LEU ref_LYS ref_MET ref_PHE ref_PRO ref_SER ref_THR ref_TRP ref_TYR ref_VAL fa_elec dslf_fa13 yhh_planarity fa_intra_sol_xover4 lk_ball_wtd rama_prepro
name                                                                                        
score12 0.8 0.44 0.6500 0.004 1.00 0.49 0.585 1.17 1.17 1.1 1.0 1.0 1.0 1.0 0.20 0.500 0.56 0.32 0.160000 -0.98000 -0.89000 -0.67000 1.700000 -0.97000 -0.81000 -0.170000 0.560000 0.240000 -0.100000 -0.650000 -0.340000 0.63000 0.020000 -0.370000 -0.27000 0.91000 0.510000 0.290000 0.000 0.00 0.000 0.0 0.0 0.00
talaris2013 0.8 0.44 0.7500 0.004 1.00 0.00 1.170 1.17 1.17 1.1 0.0 0.0 0.0 0.0 0.20 0.500 0.56 0.32 0.592942 -0.14916 -0.94198 -1.28682 0.354993 -1.17797 -1.55374 0.140526 0.357498 0.831803 0.602328 -0.287374 0.158677 0.43057 -0.219285 0.176583 0.16454 0.92933 0.131696 0.744844 0.700 1.00 0.000 0.0 0.0 0.00
talaris2014 1.0 0.55 0.9375 0.005 1.25 0.00 1.170 1.17 1.17 1.1 0.0 0.0 0.0 0.0 0.25 0.625 0.70 0.40 0.773742 -0.32436 -1.19118 -1.63002 0.443793 -1.51717 -1.96094 0.173326 0.388298 1.080600 0.761128 -0.358574 0.249477 0.61937 -0.250485 0.165383 0.20134 1.23413 0.162496 0.979644 0.875 1.25 0.625 0.0 0.0 0.00
ref2015 1.0 0.55 1.0000 0.005 1.25 0.00 1.000 1.00 1.00 1.0 0.0 0.0 0.0 0.0 0.00 0.400 0.70 0.60 1.324680 -0.09474 -1.34026 -2.14574 3.254790 -1.45095 -2.72453 0.798160 -0.300650 2.303740 1.661470 -0.714580 1.657350 1.21829 -1.643210 -0.289690 1.15175 2.26099 0.582230 2.642690 1.000 1.25 0.625 1.0 1.0 0.45
Mon, 2021-07-19 06:13
matteoferla