Weird though.In another section of the reference guide(https://rxdock.gitlab.io/documentation/devel/html/reference-guide/system-definition-file.html#sdf-cavity-mapping),the default large sphere redius is 4 A.

In its source code(https://gitlab.com/rxdock/rxdock/-/blob/master/lib/SphereSiteMapper.cxx?ref_type=heads)line 41,4A is the default.

Also,it's recommended that 10<RADIUS<20.Maybe the RADIUS is so big that it effects the performance.

Hope this helps you.

Using the default large sphere redius is enough.Too large redius makes the large sphere meaningless.

The first line of 6TRX-6TRX.prm is RBT_PARAMETER_FILE_V1.00 instead of RBT PARAMETER_FILE_V1.00.

Good luck.

A dirty way:

meson -Dbuildtype=release -Dtests=false some-xxx-builddir

I'm not sure why it is wrong,maybe gtest's system path changes too.

Rxdock didn't update for some time.

Your eigen is too new for rxdock.

Try to install libeigen3-dev 3.3.7 version according to the official install guide.

Yes.

Install eigen library first.

Usually,you also need to move eigens' library directory into its parent path.

It's a common problem when using cmake and eigen.I don't have computers nearby,but I'm sure there are lots of the same problem and solutions from search engines.

😆

Hello there!

I installed and configured fcitx5 in kali.

In esm mode,fcitx5 can autostart and shows input method status in notification.

In SL mode,fcitx5 can't do that,while fcitx5 is in process.

In archlinux/KDE,fcitx5 stays in process but doesn't work either through tigervnc.

In archlinux/KDE,without specifying .win-kex.rdp,xrdp+mstsc can't make fcitx5 works.By copying kex's rdp configuration into archlinux,archlinux/KDE/xrdp/mstsc finally makes fcitx5 work.

What's the difference between tigervnc and xrdp that cause tigervnc not working?And how to make SL mode support fcitx5 too?

This warning comes from source code RbtMOL2FileSource.cxx's Parse functon,which return Warning when the given atomtype is undifined by RbtTriposAtomType.h.

This warning will prevent rbmoegrid.cxx from generating MOE grid files for cavity visualization.If you don't use MOE grids for visualization,just ignore the warning and don't set rbcavity's -m flag.Use -d flag if you still need visualization.

This warning does not affect cavity searching or docking results since Na and K alse has the AtomType UNDEFINED.By the way,rdock added Zn's ionic parameters in RbtIonicAtoms.prm.All metals with atomic id <=30 can be calculated correctly,which means that metalloenzymes that have a catalytic Fe center and so on need to modife more source codes.

For patched-rdock version,there is no easy way to supress warnings.You may try the master branch(namely,rxdock) and use rxcmd -v -2 ........ to output only errors.But rxdock may consume some time to restart learning and compiling rxdock because rxdock uses totally different strategies of organizing source codes and compiling.

Good luck anyway.

@ikmatar

rxdock.score.intra.dihedral.0 uses the initial conformation's score as the zero point offset.

Keep using rxdock.score.intra.dihedral is ok because Rxdock uses absolute score to compute the final score.

@sunhwan1

Van der Waals parameters in Tripos 5.2 force field (R = radius (Å); K = well depth (kcal/mol); IP = Ionization potential (eV); POL = polarisability (10²⁵ cm³)).

Copied from rxdock's reference guide manual

According to the reference in Tripos52_vdw.prm,the Tripos 5.2 forcefield parameters come from J Comp Chem, 10 (1989) 982-1012,extended by GOLD parameters(J Mol Biol, 267 (1997) 727-748).

Gold paper mentions Zinc's coordination energy,but doesn't provide Zinc's vdw parameters,neither does Tripos 5.2.

In Sybyl-X 2.1,Zinc's TRIPOS force field TAFF's vdw radius is 1.676, K=0.1.

Using these parameters may be suitable,and citing Clark, M. , R. Iii , and N. V. Opdenbosch . "Validation of the general purpose tripos 5.2 force field." Journal of Computational Chemistry 10.8(2010):982–1012. may be enough.

@ikmatar

1YET.json and 1koc.json for test miss 'sections' key,leading to FileParseError by ParameterFileSource when testing.

alchemistcai (5f7d401d) at 23 Oct 10:40

*.prm files in old test sets(ASTEX_rDock_TestSet,DUD_rDock_TestSet,RNA_rDock_TestSet) cannot be parsed in branch 307d7d54.
I wrote a script to generate system defination file templates according to the files in specified directories recursively.
It can generate json files for rxcmd cavity-search -r or rxcmd dock -r in old test sets.
In my computer,several cavity-search,dock,rmsd tests were performed and no obvious wrong happened.
It can also provide some help to users who don't know how to write system defination files manually.
I didn't find which source script do this.
The default key-value pairs are inferred from the source code and test examples,which may not be completely right.
I don't know how to set the default value of segment-name to make rxdock skipping segment-name parsing.0 and false both lead to BAD_RECEPTOR_FILE error with information:File source contains zero atoms! Have you defined RECEPTOR_SEGMENT_NAME correctly?' So in this script I skippedsegment-name`.

Download the source code and copy SDRecord.pm to xxx/lib will work.

https://gitlab.com/rxdock/rxdock/-/raw/master/lib/SDRecord.pm?inline=false.

SDRecord.pm is in $RBT_ROOT/lib.Define the environment variable $RBT_ROOT and add $RBT_ROOT/lib to the environment variable LD_LIBRARY_PATH before using sdsort or other Perl scripts.

Read this website for more infomation about RxDock's environment variables:https://rxdock.gitlab.io/documentation/devel/html/reference-guide/configuration.html?highlight=environment

*.prm files in old test sets(ASTEX_rDock_TestSet,DUD_rDock_TestSet,RNA_rDock_TestSet) cannot be parsed in branch 307d7d54.
I wrote a script to generate system defination file templates according to the files in specified directories recursively.
It can generate json files for rxcmd cavity-search -r or rxcmd dock -r in old test sets.
In my computer,several cavity-search,dock,rmsd tests were performed and no obvious wrong happened.
It can also provide some help to users who don't know how to write system defination files manually.
I didn't find which source script do this.
The default key-value pairs are inferred from the source code and test examples,which may not be completely right.
I don't know how to set the default value of segment-name to make rxdock skipping segment-name parsing.0 and false both lead to BAD_RECEPTOR_FILE error with information:File source contains zero atoms! Have you defined RECEPTOR_SEGMENT_NAME correctly?' So in this script I skippedsegment-name`.

Because of the changes of score function name from SCORE.* to rxdock.score.*,sdreport can't report correctly by using the default command sdreport -l.All numbers of the output are zero.
By changing @defaultFields and @defaultedNormFields to be consistent with the main docking engine,the results turn to be correct.