i_fdZddlZddlZddlmZddlmZmZddl m Z m Z ddl m Z dd lmZmZmZmZmZmZmZmZmZmZmZmZmZmZmZmZddlZddl Z e j!d Z"Gd d eZ#Gd de Z$dS)a" AMBER PRMTOP topology parser ============================ Reads an AMBER top file to build the system. Amber keywords are turned into the following attributes: +----------------------------+----------------------+ | AMBER flag | MDAnalysis attribute | +============================+======================+ | ATOM_NAME | names | +----------------------------+----------------------+ | CHARGE | charges | +----------------------------+----------------------+ | ATOMIC_NUMBER | elements | +----------------------------+----------------------+ | MASS | masses | +----------------------------+----------------------+ | BONDS_INC_HYDROGEN | bonds | | BONDS_WITHOUT_HYDROGEN | | +----------------------------+----------------------+ | ANGLES_INC_HYDROGEN | angles | | ANGLES_WITHOUT_HYDROGEN | | +----------------------------+----------------------+ | DIHEDRALS_INC_HYDROGEN | dihedrals / improper | | DIHEDRALS_WITHOUT_HYDROGEN | | +----------------------------+----------------------+ | ATOM_TYPE_INDEX | type_indices | +----------------------------+----------------------+ | AMBER_ATOM_TYPE | types | +----------------------------+----------------------+ | RESIDUE_LABEL | resnames | +----------------------------+----------------------+ | RESIDUE_POINTER | residues | +----------------------------+----------------------+ TODO: Add support for Chamber-style topologies More stringent tests .. Note:: The Amber charge is converted to electron charges as used in MDAnalysis and other packages. To get back Amber charges, multiply by 18.2223. Chamber-style Amber topologies (i.e. topologies generated via parmed conversion of a CHARMM topology to an AMBER one) are not currently supported. Support will likely be added in future MDAnalysis releases. As of version 2.0.0, elements are no longer guessed if ATOMIC_NUMBER records are missing. In those scenarios, if elements are necessary, users will have to invoke the element guessers after parsing the topology file. Please see :mod:`MDAnalysis.guessers` for more details. .. _`PARM parameter/topology file specification`: https://ambermd.org/FileFormats.php#topo.cntrl Classes ------- .. autoclass:: TOPParser :members: :inherited-members: N)Z2SYMB)openany FORTRANReader)TopologyReaderBase change_squash)Topology) Atomnames AtomtypesAtomidsChargesElementsMassesResnamesResidsResnumsSegidsChainIDsAtomAttrBondsAngles Dihedrals ImproperszMDAnalysis.topology.TOPParserceZdZdZdZdZdZdS) TypeIndiceszNumerical type of each Atom type_indices type_indexatomN)__name__ __module__ __qualname____doc__attrnamesingularlevelg/srv/www/vhosts/g4struct/public_html/venv/lib/python3.11/site-packages/MDAnalysis/topology/TOPParser.pyrr{s#%%HH EEEr(rc~eZdZdZgdZdZdZdZdZdZ dZ d Z d Z d Z d Zd ZdZdZdZdedefdZdS) TOPParseraReads topology information from an AMBER top file. Reads the following Attributes if in topology: - Atomnames - Charges - Masses - Elements - Atomtypes - Resnames - Type_indices - Bonds - Angles - Dihedrals (inc. impropers) - ChainIDs (from %RESIDUE_CHAINID) - Segids (from %RESIDUE_CHAINID) The format is defined in `PARM parameter/topology file specification`_. The reader tries to detect if it is a newer (AMBER 12?) file format by looking for the flag "ATOMIC_NUMBER". .. _`PARM parameter/topology file specification`: https://ambermd.org/FileFormats.php#topo.cntrl Additionally, the RESIDUE_CHAINID non-standard flag is supported. This can be added with the `addPDB`_ command from parmed: .. _`addPDB`: https://parmed.github.io/ParmEd/html/parmed.html#addpdb Notes ----- Elements are obtained from the atomic numbers (if present). If a given input atomic number does not belong to an element (usually either -1 or 0), the element will be assigned an empty record. .. versionchanged:: 0.7.6 parses both amber10 and amber12 formats .. versionchanged:: 0.19.0 parses bonds, angles, dihedrals, and impropers .. versionchanged:: 1.0.0 warns users that chamber-style topologies are not currently supported .. versionchanged:: 2.0.0 no longer guesses elements if missing .. versionchanged:: 2.7.0 gets Segments and chainIDs from flag RESIDUE_CHAINID, when present )TOPPRMTOPPARM7c b ddjddfddjddfddjddfddjd dfddjd dfddjd dfddjd d fddjdd fddjddfddjddfddjddfddjddfddjddfddjddfddj dd fd}i tj 5_ tj }|ds'tdj tj }|ddksP|ddkrd j }nd!j }t||d"s)tj }|d")tj fd#t%dD}d$|D}tj }|d%d}| ||\} } } } } || | z}|| z}|| zdkr|dz }| | | | <fd'}n#t($rfd(}YnwxYw |}|dkr|s |} |d%d}n;#t*$rd)j d*}t+|d&wxYw#t,$rd&}YnwxYw|d&d&d&n #1swxYwYt/ d} d}||t5j|t4j+}t;t=|d&d,|dd&D]\}\}}||||<t/ d }t?d-tAj! d dD d.<tEd/tAj! d dD d0<# d d\ d1< d2<d vr1d3}tH%|tMj'|nSt5j( dj)d4kr0d5}tH%|tMj'|tUt5j+|dz d6<tYt5j+|dz d7<t[t5j+|dz d8<d vr~t/ d|kret] df df\}\} fd9|D}t_| d<ta| d:<t/|}n~d vrJd;|dgtd+ d<d&}d}tg|||ti )||?}|S)@z{Parse Amber PRMTOP topology file *filename*. Returns ------- A MDAnalysis Topology object rnamercharge elementsmassrtypesresname respointbondhrbondaanghangadihhdihasegids) ATOM_NAMECHARGE ATOMIC_NUMBERMASSATOM_TYPE_INDEXAMBER_ATOM_TYPE RESIDUE_LABELRESIDUE_POINTERBONDS_INC_HYDROGENBONDS_WITHOUT_HYDROGENANGLES_INC_HYDROGENANGLES_WITHOUT_HYDROGENDIHEDRALS_INC_HYDROGENDIHEDRALS_WITHOUT_HYDROGENRESIDUE_CHAINIDz%VEz2{0} is not a valid TOP file. %VE Missing in headerTITLECTITLEze{0} is detected as a Chamber-style TOP file. At this time MDAnalysis does not support such topologiesz6{0} is not a valid TOP file. 'TITLE' missing in headerz%FLAG POINTERSc\g|](}tj)Sr')nexttopfilestrip).0iselfs r) z#TOPParser.parse..s/GGG$t|,,2244GGGr(cZg|](}|D]}t|)Sr')splitint)r[r\ks r)r^z#TOPParser.parse..s3FFF1AGGIIFFqAFFFFr(%FLAGNc,tjSN)rXrYr]sr) next_getterz$TOPParser.parse..next_getters#DL111r(c,Sre)skipperrfsr)rgz$TOPParser.parse..next_getters#||~~-r(z9%FLAG section not found, formatting error for PARM7 file  dtypecg|]}|Sr'r'r[r\s r)r^z#TOPParser.parse..4s(      r(bondscg|]}|Sr'r'ros r)r^z#TOPParser.parse..=s N N N1Q N N Nr(angles dihedrals improperszATOMIC_NUMBER record not found, elements attribute will not be populated. If needed these can be guessed using universe.guess_TopologyAttrs(to_guess=['elements']).zUnknown ATOMIC_NUMBER value found for some atoms, these have been given an empty element record. If needed these can be guessed using universe.guess_TopologyAttrs(to_guess=['elements']).atomidsresidsresnumsc,g|]}d|S)rEr')r[rattrss r)r^z#TOPParser.parse..cs";;;qh*;;;r(rzNumber of residues (z-) does not match number of %RESIDUE_CHAINID (z). Skipping section.SYSTEM)r{ atom_resindexresidue_segindex)5 parse_names parse_chargesparse_elements parse_massesparse_type_indices parse_typesparse_resnames parse_residx parse_bondedparse_chainidsrfilenamerYrX startswith ValueErrorformatr`rangerZKeyError IndexError StopIterationlenpopappendnpzerosint32 enumeratezipr itertoolschainrparse_dihedralsloggerwarningwarningswarnanyvaluesr arangerrr rrarrayobjectr list)!r]kwargssectionsheadertitleemsg topremarkssys_info next_sectionnum_per_recordper_linefuncr1sect_numnumnumlinesrglineerrmsgn_atomsresptrsresidxr\xyn_resmsgsegidxrEchainidsn_segstopr{s!` @r)parsezTOPParser.parsesR!161=!T/1=T%8*aH4,fa8'  !"2t'7!DT%8)RH !2t'8*bI#$b$*;Wa"H'("d.?!&L$%r4+!> -x"U8_$6!! FIV<;;;F;;;H$VnnE(O ( 2 2E* [[FF5  T5TT),U8_)=)=TTTs### c"""$RXxj%G%G%GHHE(OFF   u||~~&& #     smFMJ$3MJ)&M(J))M-.L3-L  M &L00M3 M?MMMMMct|j}|ds)t|j}|d)|S)zTOPParser :class: helper function, skips lines of input parm7 file until we find the next %FLAG entry and return that Returns ------- line : string String containing the current line of the parm7 file rc)rXrYr)r]rs r)rizTOPParser.skippersQDL!!//'** & %%D//'** & r(c||d}ttj|t}|S)aExtracts atoms names from parm7 file Parameters ---------- num_per_record : int The number of entries for each record in the section (unused input) numlines : int The number of lines to be parsed in current section Returns ------- attr : :class:`Atomnames` A :class:`Atomnames` instance containing the names of each atom as defined in the parm7 file c|Srer'rs r)z'TOPParser.parse_names..Ar(rk)parsesection_mapperr rrrr]rrvalsattrs r)rzTOPParser.parse_names< ''++>>$f55566 r(c||d}ttj|t}|S)aExtracts the names of each residue Parameters ---------- num_per_record : int The number of entries for each recrod in section (unused input) numlines : int The number of lines to be parsed in current section Returns ------- attr : :class:`Resnames` A :class:`Resnames` instance containing the names of each residue as defined in the parm7 file c|Srer'rs r)rz*TOPParser.parse_resnames..rr(rk)rrrrrrs r)rzTOPParser.parse_resnamess< ''++>>V44455 r(c||d}tj|tj}|dz}t |}|S)aExtracts the partial charges for each atom Parameters ---------- num_per_record : int The number of entries for each record in section (unused input) numlines : int The number of lines to be parsed in current section Returns ------- attr : :class:`Charges` A :class:`Charges` instance containing the partial charges of each atom as defined in the parm7 file c t|Srefloatrs r)rz)TOPParser.parse_charges.. E!HHr(rkg82@)rrrfloat32r)r]rrrchargesrs r)rzTOPParser.parse_chargessO ''2D2DEE(4rz2227w r(cR||d}t|}|S)aExtracts the mass of each atom Parameters ---------- num_per_record : int The number of entries for each record in section (unused input) numlines : int The number of lines to be parsed in current section Returns ------- attr : :class:`Masses` A :class:`Masses` instance containing the mass of each atom as defined in the parm7 file c t|Srerrs r)rz(TOPParser.parse_masses..rr()rrrs r)rzTOPParser.parse_massess- ''2D2DEEd|| r(c||d}ttj|t}|S)aExtracts the atomic numbers of each atom and converts to element type Parameters ---------- num_per_record : int The number of entries for each record in section(unused input) numlines : int The number of lines to be pasred in current section Returns ------- attr : :class:`Elements` A :class:`Elements` instance containing the element of each atom as defined in the parm7 file Note ---- If the record contains unknown atomic numbers (e.g. <= 0), these will be treated as unknown elements and assigned an empty string value. See issues #2306 and #2651 for more details. .. versionchanged:: 2.0.0 Unrecognised elements will now return a empty string. The parser will no longer attempt to guess the element by default. c`t|dkrtt|ndS)Nrru)rarrs r)rz*TOPParser.parse_elements..s"#a&&1**s1vv"r(rk)rrrrrrs r)rzTOPParser.parse_elementssE6'' DD  V44455 r(c||d}ttj|t}|S)aExtracts the force field atom types of each atom Parameters ---------- num_per_record : int The number of entries for each record in section (unused input) numlines : int The number of lines to be parsed in current section Returns ------- attr : :class:`Atomtypes` A :class:`Atomtypes` instance containing the atom types for each atom as defined in the parm7 file c|Srer'rs r)rz'TOPParser.parse_types..rr(rk)rr rrrrs r)rzTOPParser.parse_typesrr(c||d}ttj|tj}|S)aExtracts the index of atom types of the each atom involved in Lennard Jones (6-12) interactions. Parameters ---------- num_per_record : int The number of entries for each record in section (unused input) numlines : int The number of lines to be parsed in current section Returns ------- attr :class:`TypeIndices` A :class:`TypeIndices` instance containing the LJ 6-12 atom type index for each atom c t|Srerars r)rz.TOPParser.parse_type_indices..%s CFFr(rk)rrrrrrs r)rzTOPParser.parse_type_indicess@"''2B2BCC28D999:: r(c4||d}|S)aExtracts the residue pointers for each atom Parameters ---------- num_per_record : int The number of entries for each record in section (unused input) numlines : int The number of lines to be parsed in current section Returns ------- vals : list of int A list of zero-formatted residue pointers for each atom c&t|dz S)Nrrrs r)rz(TOPParser.parse_residx..8sCFFQJr()r)r]rrrs r)rzTOPParser.parse_residx)s"''2F2FGG r(c`fdtdtD}|S)aHelper function to parse AMBER PRMTOP bonds/angles. Parameters ---------- data : list of int Input list of the parm7 bond/angle section, zero-indexed num_per_record : int The number of entries for each record in the input list Returns ------- vals : list of int tuples A list of tuples containing the atoms involved in a given bonded interaction Note ---- In the parm7 format this information is structured in the following format: [ atoms 1:n, internal index ] Where 1:n represent the ids of the n atoms involved in the bond/angle and the internal index links to a given set of FF parameters. Therefore, to extract the required information, we split out the list into chunks of size num_per_record, and only extract the atom ids. cJg|]}t||zdz  S)r)tuple)r[r chunksizedatas r)r^z*TOPParser.parse_chunks..TsD    $q1y=1,,- . .   r(r)rr)r]rrrs `` r) parse_chunkszTOPParser.parse_chunks;sJ2     1c$ii33    r(c`||d}|||}|S)aExtracts bond information from PARM7 format files Parameters ---------- num_per_record : int The number of entries for each record in section numlines : int The number of lines to be parsed for this section Note ---- For the bond/angle sections of parm7 files, the atom numbers are set to coordinate array index values. As detailed in `the specification`_, to recover the actual atom number, one should divide the values by 3 and add 1. Here, since we want to satisfy zero-indexing, we only divide by 3. .. _`the specification`: https://ambermd.org/FileFormats.php#topo.cntrl c&t|dzS)Nr;rrs r)rz(TOPParser.parse_bonded..nsc!ffkr()rr)r]rrfieldssections r)rzTOPParser.parse_bondedZs7())(4I4IJJ##FN;;r(c g} fd  |j}t|}t|D]}t|j}tt |jD]^}||j|j|j|j} | r| || _|S)aParses FORTRAN formatted section, and returns a list of all entries in each line Parameters ---------- numlines : int The number of lines to be parsed in this section mapper : lambda operator Operator to format entries in current section Returns ------- section : list A list of all entries in a given parm7 section ct|}|dddkr#|ddddS|dddkr |Std|) zZSkips '%COMMENT' lines until it gets the FORMAT specification for the section.NrDz%FORMAT)rz%COMMENTzMInvalid header line. Does not begin with either %FLAG, %FORMAT nor %COMMENT: )rXr`r)filerget_fmts r)rz.TOPParser.parsesection_mapper..get_fmts::DBQBx9$$ABBx~~c**1--bqbZ''wt}}$ -&*--r() rYrrrXrentriesstartstoprZr) r]rmapperrfmtrr\ljvalrs @r)rzTOPParser.parsesection_mapperrs      gdl## #  x 0 0AT\""A3qy>>** 0 0 ! *QYq\->>?EEGG0NN66#;;/// 0r(c,g}g}tj||D]}|ddkrL||ddt|dfzt|dfzZ|ddkrB|ddt|dfz|ddz}||||t t |}t |}t|}||fS)aCombines hydrogen and non-hydrogen containing AMBER dihedral lists and extracts sublists for conventional dihedrals and improper angles Parameters ---------- diha : list of tuples The atom ids of dihedrals not involving hydrogens dihh : list of tuples The atom ids of dihedrals involving hydrogens Returns ------- dihedrals : :class:`Dihedrals` A :class:`Dihedrals` instance containing a list of all unique dihedrals as defined by the parm7 file impropers : :class:`Impropers` A :class:`Impropers` instance containing a list of all unique improper dihedrals as defined by the parm7 file Note ---- As detailed in `the specification`_, the dihedral sections of parm7 files contain information about both conventional dihedrals and impropers. The following must be accounted for: 1) If the fourth atom in a dihedral entry is given a negative value, this indicates that it is an improper. 2) If the third atom in a dihedral entry is given a negative value, this indicates that it 1-4 NB interactions are ignored for this dihedrals. This could be due to the dihedral within a ring, or if it is part of a multi-term dihedral definition or if it is an improper. .. _`the specification`: https://ambermd.org/FileFormats.php#topo.cntrl r;rNr)rrrabssortedsetrr) r]rCrAimpropdihedr\rrsrts r)rzTOPParser.parse_dihedralss Dt,,  Ataxx aes1Q4yyl2c!A$ii\ABBBB1!uAaD |+ae3 T"""" Qs5zz""e$$ f%% )##r(rrc\||d}tj|}|S)aExtracts the chainID of each residue Parameters ---------- num_per_record : int The number of entries for each record in section (unused input) numlines : int The number of lines to be parsed in current section Returns ------- attr : numpy array A numpy array containing the chainID of each residue as defined in the parm7 file c|Srer'rs r)rz*TOPParser.parse_chainids..rr()rrrrs r)rzTOPParser.parse_chainidss- ''++>>x~~ r(N)r r!r"r#rrrirrrrrrrrrrrrrarr'r(r)r+r+s',,\( ' 'FIIIV   ((,(B(*$>0)))V/$/$/$bSCr(r+)%r#numpyrrguesser.tablesrlib.utilrrbaserr core.topologyr core.topologyattrsr r r rrrrrrrrrrrrrrlogging getLoggerrrr+r'r(r)r s0BBF######--------33333333$$$$$$&  : ; ;(] ] ] ] ] "] ] ] ] ] r(