Explicit pockets with various alternates handling

Documentation stuff missing

.dockerignore

@@ -0,0 +1 @@
+*.o

.gitignore

@@ -5,3 +5,4 @@ src/qhull/bin/
 src/qhull/lib/libqhullstatic_r.a
 src/qhull/lib/libqhullstatic.a
 *.pyc
+vmd/plugins

Dockerfile-debian-slim

@@ -1,7 +1,6 @@
-FROM centos:7
-
-#RUN yum -y install epel-release && yum -y update && yum -y install gcc gcc-c++ make netcdf-devel; yum clean all
-RUN yum -y install gcc gcc-c++ make netcdf-devel; yum clean all
+FROM debian:bullseye-slim
+RUN groupadd -r fpocket && useradd --no-log-init -r -g fpocket fpocket
+RUN apt update -y && apt install -y gcc g++ make libnetcdf-dev && rm -rf /var/lib/apt/lists/*
 
 # all of this mess is essentially to have a minimalistic build at the end
 COPY makefile /opt/fpocket/
@@ -17,8 +16,7 @@ COPY plugins/noarch /opt/fpocket/plugins/noarch
 
 WORKDIR /opt/fpocket
 
-RUN make; make install; make clean
-
-WORKDIR /WORKDIR
-
+RUN make && make install && make clean
+USER fpocket
+WORKDIR /tmp
 CMD ["fpocket"]

Dockerfile-molfile-debug

@@ -0,0 +1,29 @@
+FROM ubuntu:latest
+RUN groupadd -r fpocket && useradd --no-log-init -r -g fpocket fpocket
+ENV DEBIAN_FRONTEND=noninteractive
+ENV PLUGINDIR=compiled
+RUN apt update -y && apt install -y vim gdb gcc g++ make libnetcdf-dev && rm -rf /var/lib/apt/lists/* 
+
+# all of this mess is essentially to have a minimalistic build at the end
+COPY vmd /vmd
+WORKDIR /vmd/plugins
+RUN make LINUXAMD64 && make distrib
+
+COPY makefile /opt/fpocket/
+COPY src /opt/fpocket/src
+COPY data/sample /opt/fpocket/sample
+COPY man /opt/fpocket/man
+COPY headers /opt/fpocket/headers
+COPY obj /opt/fpocket/obj
+COPY scripts /opt/fpocket/scripts
+COPY bin /opt/fpocket/bin
+COPY plugins/LINUXAMD64 /opt/fpocket/plugins/LINUXAMD64
+COPY plugins/include /opt/fpocket/plugins/include
+COPY plugins/noarch /opt/fpocket/plugins/noarch
+
+WORKDIR /opt/fpocket
+RUN cp -r /vmd/plugins/molfile_plugin/compiled/LINUXAMD64/molfile/* plugins/LINUXAMD64/molfile/
+#RUN make && make install && make clean
+USER fpocket
+WORKDIR /tmp
+CMD ["fpocket"]

README.md

@@ -18,6 +18,8 @@ The documentation below here is just a quick & rough overview.
 
 ## What's new compared to fpocket 2.0 (old sourceforge repo)
 __fpocket__: 
+- fpocket now supports mmCIF as input and output format together with the classical PDB format
+- support was added to define / delete and handle protein chains or sets of them to enable characterization of protein protein binding epitopes
 - is now able to consider explicit pockets when you want to calculate properties for a known binding site
 - cli changed a bit
 - pocket flexibility using temperature factors is better considered (less very flexible pockets on very solvent exposed areas)
@@ -152,6 +154,11 @@ You can run fpocket using the following command line as an example:
 fpocket -f 1uyd.pdb
 ```
 
+fpocket now also eats cif as input, so this would work as well. Make sure to use proper file extensions
+```bash
+fpocket -f 1uyd.cif
+```
+
 This will detect all pockets on the input pdb file, named 1uyd.pdb
 If you want to get all command line args for fpocket, simply type `fpocket``
 
@@ -174,7 +181,7 @@ Please read [CONTRIBUTING.md](https://gist.github.com/PurpleBooth/b24679402957c6
 
 * **Peter Schmidtke** - *Initial work* - [pschmidtke](https://github.com/pschmidtke)
 * **Vincent Le Guilloux** - *Initial work* - [leguilv](https://github.com/leguilv)
-* **Mael Shorkar** - *Chain handling* - [leguilv](https://github.com/shorkarmael)
+* **Mael Shorkar** - *Chain handling, MMCIF support* - [shorkarmael](https://github.com/shorkarmael)
 
 
 ## License

data/sample/5RGF.cif

@@ -3,7 +3,7 @@ data_5RGF
 _entry.id   5RGF 
 # 
 _audit_conform.dict_name       mmcif_pdbx.dic 
-_audit_conform.dict_version    5.330 
+_audit_conform.dict_version    5.342 
 _audit_conform.dict_location   http://mmcif.pdb.org/dictionaries/ascii/mmcif_pdbx.dic 
 # 
 loop_
@@ -36,18 +36,19 @@ _audit_author.identifier_ORCID
 'Chica, R.A.'         5 ? 
 # 
 _citation.id                        primary 
-_citation.title                     'Evolution of an enzyme conformational ensembles guides the design of an efficient biocatalyst' 
-_citation.journal_abbrev            'To be published' 
-_citation.journal_volume            ? 
-_citation.page_first                ? 
-_citation.page_last                 ? 
-_citation.year                      ? 
-_citation.pdbx_database_id_PubMed   ? 
-_citation.pdbx_database_id_DOI      ? 
+_citation.title                     
+'Ensemble-based enzyme design can recapitulate the effects of laboratory directed evolution in silico.' 
+_citation.journal_abbrev            'Nat Commun' 
+_citation.journal_volume            11 
+_citation.page_first                4808 
+_citation.page_last                 4808 
+_citation.year                      2020 
+_citation.pdbx_database_id_PubMed   32968058 
+_citation.pdbx_database_id_DOI      10.1038/s41467-020-18619-x 
 _citation.journal_id_ASTM           ? 
-_citation.country                   ? 
-_citation.journal_id_ISSN           ? 
-_citation.journal_id_CSD            0353 
+_citation.country                   UK 
+_citation.journal_id_ISSN           2041-1723 
+_citation.journal_id_CSD            ? 
 _citation.book_publisher            ? 
 # 
 loop_
@@ -55,11 +56,15 @@ _citation_author.citation_id
 _citation_author.name 
 _citation_author.identifier_ORCID 
 _citation_author.ordinal 
-primary 'Broom, A.'           ? 1 
-primary 'Rakotoharisoa, R.V.' ? 2 
-primary 'Thompson, M.'        ? 3 
-primary 'Fraser, J.'          ? 4 
-primary 'Chica, R.A.'         ? 5 
+primary 'Broom, A.'           ?                   1 
+primary 'Rakotoharisoa, R.V.' ?                   2 
+primary 'Thompson, M.C.'      0000-0002-6099-2027 3 
+primary 'Zarifi, N.'          0000-0003-4748-7082 4 
+primary 'Nguyen, E.'          ?                   5 
+primary 'Mukhametzhanov, N.'  ?                   6 
+primary 'Liu, L.'             0000-0003-1283-1410 7 
+primary 'Fraser, J.S.'        0000-0002-5080-2859 8 
+primary 'Chica, R.A.'         0000-0003-3789-9841 9 
 # 
 _cell.entry_id           5RGF 
 _cell.length_a           76.410 
@@ -18221,11 +18226,15 @@ _pdbx_struct_oper_list.matrix[3][2]         0.0000000000
 _pdbx_struct_oper_list.matrix[3][3]         1.0000000000 
 _pdbx_struct_oper_list.vector[3]            0.0000000000 
 # 
-_pdbx_audit_revision_history.ordinal             1 
-_pdbx_audit_revision_history.data_content_type   'Structure model' 
-_pdbx_audit_revision_history.major_revision      1 
-_pdbx_audit_revision_history.minor_revision      0 
-_pdbx_audit_revision_history.revision_date       2020-07-22 
+loop_
+_pdbx_audit_revision_history.ordinal 
+_pdbx_audit_revision_history.data_content_type 
+_pdbx_audit_revision_history.major_revision 
+_pdbx_audit_revision_history.minor_revision 
+_pdbx_audit_revision_history.revision_date 
+1 'Structure model' 1 0 2020-07-22 
+2 'Structure model' 1 1 2020-12-02 
+3 'Structure model' 1 2 2021-05-12 
 # 
 _pdbx_audit_revision_details.ordinal             1 
 _pdbx_audit_revision_details.revision_ordinal    1 
@@ -18236,6 +18245,41 @@ _pdbx_audit_revision_details.description         ?
 _pdbx_audit_revision_details.details             ? 
 # 
 loop_
+_pdbx_audit_revision_group.ordinal 
+_pdbx_audit_revision_group.revision_ordinal 
+_pdbx_audit_revision_group.data_content_type 
+_pdbx_audit_revision_group.group 
+1 2 'Structure model' 'Database references' 
+2 3 'Structure model' 'Structure summary'   
+# 
+loop_
+_pdbx_audit_revision_category.ordinal 
+_pdbx_audit_revision_category.revision_ordinal 
+_pdbx_audit_revision_category.data_content_type 
+_pdbx_audit_revision_category.category 
+1 2 'Structure model' citation           
+2 2 'Structure model' citation_author    
+3 3 'Structure model' pdbx_deposit_group 
+# 
+loop_
+_pdbx_audit_revision_item.ordinal 
+_pdbx_audit_revision_item.revision_ordinal 
+_pdbx_audit_revision_item.data_content_type 
+_pdbx_audit_revision_item.item 
+1  2 'Structure model' '_citation.country'                 
+2  2 'Structure model' '_citation.journal_abbrev'          
+3  2 'Structure model' '_citation.journal_id_CSD'          
+4  2 'Structure model' '_citation.journal_id_ISSN'         
+5  2 'Structure model' '_citation.journal_volume'          
+6  2 'Structure model' '_citation.page_first'              
+7  2 'Structure model' '_citation.page_last'               
+8  2 'Structure model' '_citation.pdbx_database_id_DOI'    
+9  2 'Structure model' '_citation.pdbx_database_id_PubMed' 
+10 2 'Structure model' '_citation.title'                   
+11 2 'Structure model' '_citation.year'                    
+12 3 'Structure model' '_pdbx_deposit_group.group_type'    
+# 
+loop_
 _pdbx_refine_tls.pdbx_refine_id 
 _pdbx_refine_tls.id 
 _pdbx_refine_tls.details 
@@ -18489,6 +18533,7 @@ _pdbx_audit_support.ordinal                1
 _pdbx_deposit_group.group_title         'Crystal structures of HG-series of Kemp Eliminases at Room-temperature' 
 _pdbx_deposit_group.group_description   'Crystal structure of Kemp eliminase HG4 with bound transition state analogue, 277K' 
 _pdbx_deposit_group.group_id            G_1002148 
+_pdbx_deposit_group.group_type          undefined 
 # 
 _pdbx_entity_instance_feature.ordinal        1 
 _pdbx_entity_instance_feature.comp_id        6NT 

doc/GETTINGSTARTED.md

@@ -417,7 +417,7 @@ The simplest way to run fpocket is either by providing a single pdb file, or by
 
 #### Optional:
 
-    -m float: (default 3.4Å) This flag enables the user to modify the minimum radius an alpha sphere might have in a binding pocket. An alpha sphere is a contact sphere, that touches 4 atoms in 3D space without having any internal atoms. Here 3Å allow filtering of too small (protein internal) alpha spheres. I you want to analyze internal interstices, lower this parameter. In the contrary, if you want to analyze more solvent exposed cavities, you can raise this parameter in order to filter out too buried cavities.
+    -m float: (default 3.4Å) This flag enables the user to modify the minimum radius an alpha sphere might have in a binding pocket. An alpha sphere is a contact sphere, that touches 4 atoms in 3D space without having any internal atoms. Here 3Å allow filtering of too small (protein internal) alpha spheres. If you want to analyze internal interstices, lower this parameter. In the contrary, if you want to analyze more solvent exposed cavities, you can raise this parameter in order to filter out too buried cavities.
     
     -M float: (default 6.2Å) Here you can modify the maximum radius of alpha spheres in a pocket. An alpha sphere is a contact sphere, that touches 4 atoms in 3D space without having any internal atoms. Here 7Å allow to filter out too large contact spheres, that are lying on the protein surface. If you want to analyze very flat and solvent exposed surface depressions, raise this parameter. For analysis of buried parts of the protein you can lower this parameter. Higher radii might be more interesting for identification of protein protein binding sites or polysaccharide binding sites. Smaller radii enable detection of buried cavities for small organic molecules (drugs, for instance).
     
@@ -449,13 +449,13 @@ The simplest way to run fpocket is either by providing a single pdb file, or by
     
     -v int: (default 2500) By default, pockets volume are calculated using a monte-carlo algorithm. Basically, the algorithm picks a random point in the space and check if it is included in any alpha sphere, and stores this status. This is repeated N times, and we estimate the volume of the pocket using ratio between the number of hit and the number of iteration, scaled by the size of the box. This parameter defines the number of iteration to perform. Of course, the higher the value is, the greater the accuracy will be, but the performance will be slowed down.
     
-    -b (none): (NOT USED BY DEFAULT) This option allows the user to chose a discrete algorithm  to calculate  the  volume of each pocket instead of the Monte Carlo method. This algorithm puts each pocket into a grid of  dimention (1/N*X  ;  1/N*Y  ;  1/N*Z),  N being the value given using this option, and X, Y and Z being the box dimensions,  determined using coordinates of vertices. Then, a triple iteration on each dimensions is used to estimate the volume, checking if each points  given  by the iteration is in one of the pocket’s vertices. This parameter defines the grid discretization. If this parameter is used, this algorithm will be used instead of the Monte Carlo algorithm. 
+    -b (none): (NOT USED BY DEFAULT) This option allows the user to choose a discrete algorithm  to calculate  the  volume of each pocket instead of the Monte Carlo method. This algorithm puts each pocket into a grid of dimension (1/N*X  ;  1/N*Y  ;  1/N*Z),  N being the value given using this option, and X, Y and Z being the box dimensions,  determined using coordinates of vertices. Then, a triple iteration on each dimensions is used to estimate the volume, checking if each points  given  by the iteration is in one of the pocket’s vertices. This parameter defines the grid discretization. If this parameter is used, this algorithm will be used instead of the Monte Carlo algorithm. 
     Warning: Although this algorithm could be more accurate, a high value might dramatically slow down the program, as this algorithm has a maximum complexity of N*N*N*nb_vertices, and a minimum of N*N*N !!!
 
 
     -d (none): Option allowing you to output pockets and properties in a condensed format. This will put to the stdout pocket properties in a tab separated string and write pocket files in a subfolder
     
-    -r string: (None) This parameter allows you to run fpocket in a restricted mode. Let's suppose you have a very shallow or large pocket with a ligand inside and the automatic pocket prediction always splits up you pocket or you have only a part of the pocket found. Specifying your ligand residue with -r allows you to detect and characterize you ligand binding site explicitely. For instance for `1UYD.pdb` you can specify `-r 1224:PU8:A` (residue number of the ligand: residue name of the ligand: chain of the ligand)
+    -r string: (None) This parameter allows you to run fpocket in a restricted mode. Let's suppose you have a very shallow or large pocket with a ligand inside and the automatic pocket prediction always splits up you pocket or you have only a part of the pocket found. Specifying your ligand residue with -r allows you to detect and characterize you ligand binding site explicitly. For instance for `1UYD.pdb` you can specify `-r 1224:PU8:A` (residue number of the ligand: residue name of the ligand: chain of the ligand)
     
     -y string: (filename) EXPERIMENTAL: here you can specify a topology filename in the Amber prmtop format. This can then be used by fpocket & mdpocket to calculate energy grids for your pockets. NB: you have to specify the -x flag to run energy calculations
     

headers/descriptors.h

@@ -24,6 +24,7 @@ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLI
 #include "atom.h"
 #include "aa.h"
 #include "utils.h"
+//#include "asa.h"
 
 
 /* --------------------------------STRUCTURES---------------------------------*/
@@ -88,6 +89,10 @@ typedef struct s_desc
 
 /* ------------------------------PROTOTYPES---------------------------------- */
 
+//explicit definition from asa.h 
+void set_ASA(s_desc *desc,s_pdb *pdb, s_vvertice **tvert,int nvert);
+
+
 s_desc* allocate_s_desc(void) ;
 void reset_desc(s_desc *desc) ;
 

headers/pertable.h

@@ -28,6 +28,8 @@ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLI
 float pte_get_vdw_ray(const char *symbol) ;
 float pte_get_mass(const char *symbol) ;
 float pte_get_enegativity(const char *symbol) ;
+float pte_get_enegativity_from_number(int atomicnumber) ;
+char *pte_get_element_from_number(int atomicnumber);
 int is_valid_element(const char *str, int ignore_case) ;
 int element_in_std_res(char *res_name);
 int element_in_nucl_acid(char *res_name);

headers/rpdb.h

@@ -81,7 +81,8 @@ typedef struct s_pdb
     int n_xlig_atoms;   /**number of atoms in xlig array ( number of atoms of selected atom*/
     int natoms,			/**< Number of atoms */
             nhetatm,		/**< Number of HETATM */
-            natm_lig ;		/**< Number of ligand atoms */
+            natm_lig,       /**< Number of ligand atoms */
+            natoms_h ;		/**< Number of hydrogen atoms */
 
     float A, B, C, 			/**< Side lengths of the unit cell */
           alpha, beta, gamma ;	/**< Angle between B and C, A and C, A and C */
@@ -89,6 +90,8 @@ typedef struct s_pdb
     char header[M_PDB_BUF_LEN] ; /**< Header container*/
     char fname[M_PDB_FILE_NAME_LEN];    /**< File name container*/
     float avg_bfactor;  /**<overall average B factor*/
+    int model_flag; /**<flag set to 1 if a particular model / conformation was read*/
+    int model_number; /**<number of the model read in*/
 
 } s_pdb ;
 

makefile

@@ -1,6 +1,7 @@
 #------------------------------------------------------------
 # Makefile.
 #------------------------------------------------------------
+#ARCH	    = MACOSXX86_64
 ARCH	    = LINUXAMD64
 export ARCH
 
@@ -12,8 +13,11 @@ PATH_HEADER = headers/
 PATH_BIN    = bin/
 PATH_MAN    = man/
 PATH_QHULL  = src/qhull/src
+PATH_LIB	= $(PLUGINDIR)/$(ARCH)/molfile/
 
 BINDIR  = /usr/local/bin/
+LIBDIR  = /usr/local/lib/
+
 MANDIR  = /usr/local/man/man8/
 
 
@@ -48,7 +52,8 @@ QCFLAGS     = -O -g -pg -ansi
 
 LGSL        = -L$(PATH_GSL)lib -lgsl -lgslcblas 
 LFLAGS	    = -lm -L$(PLUGINDIR)/$(ARCH)/molfile $(PLUGINDIR)/$(ARCH)/molfile/libmolfile_plugin.a -lstdc++
-NETCDFFLAGS	= -lnetcdf
+NETCDFFLAGS	= 
+#-lnetcdf
 
 #
 #------------------------------------------------------------
@@ -180,6 +185,7 @@ install:
 	cp $(PATH_BIN)$(DPOCKET) $(BINDIR)
 	cp $(PATH_BIN)$(MDPOCKET) $(BINDIR)
 	cp $(PATH_MAN)* $(MANDIR)
+	cp $(PATH_LIB)* $(LIBDIR)
 
 check:
 	./$(PATH_BIN)$(CHECK)

plugins/LINUXAMD64/molfile/libmolfile_plugin.h

@@ -207,6 +207,9 @@ extern int molfile_maeffplugin_fini(void);
 extern int molfile_orcaplugin_init(void);
 extern int molfile_orcaplugin_register(void *, vmdplugin_register_cb);
 extern int molfile_orcaplugin_fini(void);
+extern int molfile_qcschemaplugin_init(void);
+extern int molfile_qcschemaplugin_register(void *, vmdplugin_register_cb);
+extern int molfile_qcschemaplugin_fini(void);
 
 #define MOLFILE_INIT_ALL \
     molfile_abinitplugin_init(); \
@@ -276,6 +279,7 @@ extern int molfile_orcaplugin_fini(void);
     molfile_dtrplugin_init(); \
     molfile_maeffplugin_init(); \
     molfile_orcaplugin_init(); \
+    molfile_qcschemaplugin_init(); \
 
 #define MOLFILE_REGISTER_ALL(v, cb) \
     molfile_abinitplugin_register(v, cb); \
@@ -345,6 +349,7 @@ extern int molfile_orcaplugin_fini(void);
     molfile_dtrplugin_register(v, cb); \
     molfile_maeffplugin_register(v, cb); \
     molfile_orcaplugin_register(v, cb); \
+    molfile_qcschemaplugin_register(v, cb); \
 
 #define MOLFILE_FINI_ALL \
     molfile_abinitplugin_fini(); \
@@ -414,6 +419,7 @@ extern int molfile_orcaplugin_fini(void);
     molfile_dtrplugin_fini(); \
     molfile_maeffplugin_fini(); \
     molfile_orcaplugin_fini(); \
+    molfile_qcschemaplugin_fini(); \
 
 #ifdef __cplusplus
 }