[Reference for KCOMBU]
Kawabata T. Build-up algorithm for atomic correspondence between chemical structure.
J.Chem.Info.Model., 2011, 51, 1775-1787.
The source code of the kcombu is mainly written in C, and developed and executed the linux environment. Some additional programs for 2D graphics and statistical analyses are written in python. For the installation, you need the gcc compiler. If you want to use another compipler, please change the "Makefile" in the "src" directory. The standard installation procedures are as follows:
tar zxvf kcombu-src-[date].tar.gz
cd src
make -f Makefile.lkcombu
If the sources is successfully compiled, an execution file "lkcombu" will appear in the "../src" directory.
The program 'lkcombu' has two main modes for calculation: query-library search ("-M S") and all-vs-all comparison("-M A"). The simplest usages are summarized as follows:
>for query-library search $lkcombu -M S -Q [query_mol] -ifl [files of lib multi mols] -fL [S|P|2] -osc [output_file] $lkcombu -M S -Q [query_mol] -idl [dir of lib mols] -fL [S|P|2] -osc [output_file] $lkcombu -M S -Q [query_mol] -idml [dir of lib multi mols] -fL [S|P|2] -osc [output_file] >for all-vs-all comparison $lkcombu -M A -ifl [file of lib. multi mols] -fL [S|P|2] -osm [output_file] $lkcombu -M A -idl [dir of lib. mols] -fL [S|P|2] -osm [output_file] $lkcombu -M A -idml [dir of lib. multi mols] -fL [S|P|2] -osm [output_file] (-fL : file formats.'P'db,'S'df,'K'cf, '2':MOL2) >for comparison bwn two groups $lkcombu -M G -idlq [dir of query lib. mols] -idl [dir of lib mols] -fQ [S|P|2] -fL [S|P|2] -osm [output_file] >for randomly selecting library $lkcombu -M R -idl [dir of lib. mols] -nlr [num_select_mol] -tam [threMCS] -oms [out_multi_mol] -omsd [out_mol_dir] >for showing options in detail $lkcombu -h >for showing options for MCS $lkcombu -hmcs
Both modes ("-M S" and "-M A") requires library molecular files, which can be specified by the following options:
We prepare two ways to specify a molecular library.-ifl : input files for all the library molecules (file1:file2:..) [] -idl : input directory for library molecules [.] -ill : input list of library molecules[] -ihl : required header string of input file (combined with -idl option) [] -fL : file formats.'P'db,'S'df,'K'cf, '2':MOL2 [-] -aL : AtomHetero type. 'A'tom 'H'etatm 'B'oth[B]
-ifl [multi SDF file] -fL [S] -ifl [multi PDB file] -fL [P] -ifl [multi MOL2 file] -fL [2]Multiple library files can be acceptable by spliting colon symbols ':' :
-ifl [file1]:[file2]:[file3]
-idl [directory_for_library_molecule]
The program "lkcombu" can read several file formats of chemical structures: sdf, mol2, pdb, kcf and smi. The file format is distinguished by file exetentions : *.sdf -> SDF, *.mol2 ->MOL2, *.pdb -> PDB, *.kcf -> KCF, *.smi ->SMILES. For molecular files without proper file extensions, users should directly assign the format types by the option '-fQ'. For example, if a file "hoge" is for query and in SDF format, users should add the option "-Q hoge -fQ S".-Q : query molecule Q (*.sdf|*.mol2|*.pdb|*.kcf|*.smi)[] -fQ : file formats.'P'db,'S'df,'K'cf, '2':MOL2,'M':SMILES [-] -aQ : AtomHetero type. 'A'tom 'H'etatm 'B'oth[B]
-tao : Tanimoto threshold value for one atom descriptor (filtering) [0.000000] -tam : Tanimoto threshold value for MCS (output) [0.000000] -osc : output search result file [sch.out]The options "-tao" and "-tam" specify the theshold tanimoto similarity score for the search. The option "-tao" is more important. The option "-tao" specifies threshold tanimoto values for one atom descriptor, in other words, molecular formula. If a user specifies high value for this threshold (such as "-tao 0.9"), the program quickly throws away most of library molecules without calculating their MCSs. The option "-tam" specifies threshold tanimoto values for MCS; library molecule with more than "-tam" score are finally listed. The option "-osc" specifies the output file name of library search. The default file name is "sch.out". Its file format is described in the last section .
For the all-vs-all comparison ("-M A") and the two-group comparison, we prepare four types of output files as follows:
Their file formats are described in the last section .-osm : output file for similarities [sim.out] -fsm : file formats for osm. 'L':simple pairwise list.'N':list with molecular names : 'D':distance matrix. 'S'imilarity matrix [N]
The threshold score options -tao and -tam also work for the '-M A' and '-M G'. If you want focus on only similar molecular pairs ,for example,with tanimoto similarity > 0.5, adding the option '-tao 0.5' largely reduces your computation time.
-tao : Tanimoto threshold value for one atom descriptor (filtering) [0.000000] -tam : Tanimoto threshold value for MCS (output) [0.000000]
The all-vs-all comparison requires a large computation time, proportional to squares of number of library molecules. If you compare more than 1000 molecules, I recommend to use multiple CPU cores for the all-vs-all comparison, using '-div' options.
-div : Job division. {0..Ndivision-1}/(Ndivision). [0/1]
For example, if you calcualte the all-vs-all comparison using 4 CPU cores, following four commands should be executed:
lkcombu -M A -ifl [libfile] -div 0/4 -osm 0.out & lkcombu -M A -ifl [libfile] -div 1/4 -osm 1.out & lkcombu -M A -ifl [libfile] -div 2/4 -osm 2.out & lkcombu -M A -ifl [libfile] -div 3/4 -osm 3.out &After finishing the calculation, you can get the all-vs-all comparison file just by merging these four file (cat 0.out 1.out 2.out 3.out > all.out). Please note that the option '-osl' for output file should be used when the '-div' option is employed. The option '-odm' and '-osm' will generate imcomplete file (parts of scores are zero) using the '-div' option.
The options for calculating MCS are identical for those for pkcombu. See the README_pkcombu.html for the MCS options.
The result file library search ("-osc") is mainly composed of two parts: [SIMILAR_MOLECULAR_LIST] and [ATOM_MATCHING]. The part [SIMILAR_MOLECULAR_LIST] is the list of similar molecules with score (MCS and Descriptor), number of matched atom pairs, number of heavy atoms, heuristic selection score and molecular formula. The part [ATOM_MATCHING] is the list of atom number pairs, where the program made atomic correspondences. A following file is a result of searching for the query molecule G39.sdf(tamiflu) against the molecules stored under the directory '/DB/LIGAND-EXPO/SDF2d' using the tanimoto descriptor score threshold = 0.6, by TD-MCS1:
#COMMAND lkcombu -M S -Q G39.sdf -idl /DB/LIGAND-EXPO/SDF2d -con T -mtd 1 -tao 0.6 -tam 0.6 -osc G39.sch #DATE_START Jan 31,2012 15:12:12 #DATE_END Jan 31,2012 15:12:40 #COMP_TIME 27.547554 seconds #QUERY_FILENAME G39.sdf #QUERY_NAME /DB/LIGAND-EXPO/SDF/G39 #QUERY_FILETYPE S #QUERY_MOLFORMULA C14_N2_O4 #QUERY_NATOM 20 #QUERY_NHEAVYATOM 20 #ALGORITHM B #CONNECT_GRAPH T #MAX_DIF_TOPODIS 1 #THRE_TANIMOTO_MCS 0.600000 #THRE_TANIMOTO_ATOMPAIR -1.000000 #THRE_TANIMOTO_ONEATOM 0.600000 #LIBRARY_DIRECTORY /DB/LIGAND-EXPO/SDF2d #LIBRARY_DESCRIPTOR #NMOL_IN_LIBRARY 13582 #NMOL_OVER_THRE 19 #N_LIBRARY_FILE 0 [SIMILAR_MOLECULE_LIST] #[rank(1)] [molecular_num(2)] [file_num(3)] [file_offset(4)] [molecular_name(5)] #[similarity MCS(6)] [similarity ONEATOM(7)] [similarity ATOMPAIR(8)] #[Natompair(9)] [Nheavyatom(10)] [select_dis(11)] [molecular formula(12)] # [element_for_query]:COOCCCCNCOCCCOCCCCCN # [ring_for_query]: aaaa aa # [atom_number_for_query]: 11111111112 # :12345678901234567890 #[rk][num][name] [sMCS][sONE][sPAIR][Npair][Nhvyatm][seldis][molform] :-------------------- 1 9347 0 0 G39 1.000 1.000 0.000 20 20 0.000 C14_N2_O4 COOCCCCNCOCCCOCCCCCN 2 5358 0 0 FDI 0.783 0.783 0.000 18 21 31.000 C15_N2_O4 COOCCCCNCOCCC-CCCCC- 3 339 0 0 G28 0.708 0.783 0.000 17 21 14.000 C13_N3_O5 COOCCCCNCOCC---CCCCN 4 8973 0 0 ST3 0.700 0.700 0.000 14 14 16.000 C9_N2_O3 COOCCCCNCOCCC------N 5 6283 0 0 HTM 0.696 0.696 0.000 16 19 19.000 C11_N_O7 COOCCCCNCOCC-OC-CC-- 6 2501 0 0 IEM 0.696 0.773 0.000 16 19 18.000 C12_N_O6 COOCCCCNCOCC-OCC--C- 7 13532 0 0 DYM 0.667 0.667 0.000 16 20 22.000 C11_N_O8 COOCCCCNCOCC-OCC--C- 8 1534 0 0 ST2 0.667 0.667 0.000 14 15 7.000 C9_N2_O4 COOCCCCNCOCCC------N 9 4030 0 0 ST5 0.652 0.652 0.000 15 18 21.000 C11_N2_O5 COOCCCCNCOCCC-C-C--- 10 355 0 0 ST6 0.652 0.727 0.000 15 18 21.000 C11_N3_O4 COOCCCCNCOCCC-C-C--- 11 6386 0 0 TYZ 0.650 0.650 0.000 13 13 27.000 C9_N_O3 COOCCCCNCOCCC------- 12 12195 0 0 MMR 0.625 0.773 0.000 15 19 26.000 C12_N_O6 C-OC-CCNCOCCCOCCC--- 13 10041 0 0 TNY 0.615 0.750 0.000 16 22 42.000 C12_N2_O8 C-OCCCCNCOCC-OCCC-C- 14 3403 0 0 0AI 0.609 0.609 0.000 14 17 34.000 C9_N_O7 COOCCCCNCOCC-OC----- 15 10566 0 0 ST4 0.609 0.682 0.000 14 17 21.000 C10_N4_O3 COOCCCCNCOCCC-C----- 16 10077 0 0 49A 0.600 0.667 0.000 15 20 12.000 C11_N3_O6 COOCCCCNCOCC--C-C--N 17 12296 0 0 4AM 0.600 0.667 0.000 15 20 12.000 C11_N2_O7 COOCCCCNCOCC--C-C--N 18 5794 0 0 AMU 0.600 0.667 0.000 15 20 30.000 C11_N_O8 C-OC-CCNCOCCCOCCC--- 19 666 0 0 MUB 0.600 0.667 0.000 15 20 30.000 C11_N_O8 C-OC-CCNCOCCCOCCC--- [ATOM_MATCHING] 1 |1 1|2 2|3 3|4 4|5 5|6 6|7 7|8 8|9 9|10 10|11 11|12 12|13 13|14 14|15 15|16 16|17 17|18 18|19 19|20 20 2 |1 7|2 8|3 9|4 5|5 6|6 1|7 2|8 18|9 19|10 21|11 20|12 3|13 4|15 12|16 16|17 14|18 15|19 17 3 |1 1|2 2|3 3|4 4|5 5|6 6|7 7|8 8|9 9|10 10|11 11|12 12|16 19|17 17|18 18|19 20|20 21 4 |1 1|2 2|3 3|4 4|5 5|6 6|7 8|8 9|9 10|10 11|11 12|12 13|13 14|20 7 5 |1 2|2 1|3 14|4 3|5 4|6 5|7 6|8 13|9 11|10 19|11 12|12 7|14 17|15 8|17 9|18 10 6 |1 1|2 15|3 16|4 2|5 3|6 4|7 5|8 14|9 12|10 19|11 13|12 6|14 7|15 8|16 9|19 11 7 |1 1|2 13|3 14|4 2|5 3|6 4|7 5|8 12|9 10|10 20|11 11|12 6|14 17|15 7|16 8|19 9 8 |1 1|2 2|3 3|4 4|5 5|6 6|7 8|8 9|9 10|10 11|11 12|12 13|13 15|20 7 9 |1 1|2 2|3 3|4 4|5 18|6 17|7 12|8 13|9 14|10 15|11 16|12 6|13 5|15 8|17 10 10 |1 1|2 2|3 3|4 4|5 18|6 17|7 12|8 13|9 14|10 15|11 16|12 6|13 5|15 8|17 10 11 |1 3|2 1|3 2|4 4|5 9|6 8|7 7|8 11|9 10|10 12|11 13|12 6|13 5 12 |1 6|3 11|4 5|6 1|7 2|8 12|9 13|10 15|11 14|12 3|13 4|14 9|15 16|16 17|17 18 13 |1 13|3 14|4 11|5 9|6 7|7 5|8 6|9 15|10 16|11 17|12 4|14 20|15 19|16 18|17 21|19 1 14 |1 2|2 1|3 17|4 5|5 15|6 13|7 8|8 9|9 10|10 11|11 12|12 4|14 6|15 7 15 |1 1|2 2|3 3|4 4|5 17|6 16|7 11|8 12|9 13|10 14|11 15|12 6|13 5|15 8 16 |1 1|2 13|3 14|4 2|5 3|6 4|7 5|8 12|9 10|10 20|11 11|12 6|15 8|17 9|20 15 17 |1 1|2 2|3 3|4 4|5 5|6 6|7 8|8 9|9 10|10 11|11 12|12 13|15 17|17 19|20 7 18 |1 6|3 16|4 5|6 1|7 2|8 20|9 7|10 17|11 8|12 3|13 4|14 13|15 9|16 10|17 11 19 |1 6|3 16|4 5|6 1|7 2|8 20|9 7|10 17|11 8|12 3|13 4|14 13|15 9|16 10|17 11
#COMMAND lkcombu -M A -idl src/sample_sdf -con T -mtd 1 -fsm S -osm smat.out #START_DATE Feb 10,2012 9:32:14 #END_DATE Feb 10,2012 9:32:14 5 SIA 1.000 0.826 0.952 0.760 0.519 EQP 0.826 1.000 0.864 0.692 0.464 DAN 0.952 0.864 1.000 0.792 0.538 ZMR 0.760 0.692 0.792 1.000 0.483 G39 0.519 0.464 0.538 0.483 1.000
#COMMAND lkcombu -M A -idl src/sample_sdf -con T -mtd 1 -fsm D -osm dmat.out #START_DATE Feb 10,2012 9:31:59 #END_DATE Feb 10,2012 9:32:0 5 SIA 0.000 0.174 0.048 0.240 0.481 EQP 0.174 0.000 0.136 0.308 0.536 DAN 0.048 0.136 0.000 0.208 0.462 ZMR 0.240 0.308 0.208 0.000 0.517 G39 0.481 0.536 0.462 0.517 0.000
#>>Similarities in List format<< #COMMAND lkcombu -M A -idl src/sample_sdf -con T -mtd 1 -fsm L -osm list.out #DATE_START Feb 10,2012 9:32:48 #LIBRARY_DIRECTORY src/sample_sdf #NMOL_IN_LIBRARY 5 #DATA 0 SIA #DATA 1 EQP #DATA 2 DAN #DATA 3 ZMR #DATA 4 G39 0 1 0.826087 0 2 0.952381 0 3 0.760000 0 4 0.518519 1 2 0.863636 1 3 0.692308 1 4 0.464286 2 3 0.791667 2 4 0.538462 3 4 0.482759
#>>Similarities in List format<< #COMMAND lkcombu -M A -idl src/sample_sdf -con T -mtd 1 -fsm N -osm list_with_name.out #DATE_START Feb 10,2012 9:33:0 #LIBRARY_DIRECTORY src/sample_sdf #NMOL_IN_LIBRARY 5 #DATA 0 SIA #DATA 1 EQP #DATA 2 DAN #DATA 3 ZMR #DATA 4 G39 0 SIA 1 EQP 0.826087 0 SIA 2 DAN 0.952381 0 SIA 3 ZMR 0.760000 0 SIA 4 G39 0.518519 1 EQP 2 DAN 0.863636 1 EQP 3 ZMR 0.692308 1 EQP 4 G39 0.464286 2 DAN 3 ZMR 0.791667 2 DAN 4 G39 0.538462 3 ZMR 4 G39 0.482759