Kpax User Manual - version 5.0.0


What is Kpax?

Kpax is a fast protein structure database search and alignment program. It uses Gaussian functions to score very rapidly the local and spatial environment of each amino acid residue in a protein, and it uses dynamic programming to find the optimal global alignment of two proteins according to their Gaussian similarity scores. This approach allows very fast searches of structural databases, and it allows three-dimensional superpositions of proteins to be calculated rapidly. Kpax can also align and superpose unstructured peptides and peptide fragments. Additionally, Kpax version 5.0.0 can perform flexible structure alignments, multiple structure alignments, and multiple flexible structure alignments, and it can score alignments calculated by other programs, provided they are defined in a PIR or FASTA format alignment file. These days, many programs can superpose similar pairs of protein structures relatively easily. Still, only very few can perform multiple flexible structure alignments. Another strength of Kpax is its great speed and robustness, which makes it especially well suited for searching structural databases such as CATH and SCOP. Kpax may also be used to build and search a user-defined database of protein structures.

How Does it Work?

Kpax makes an initial structure-based alignment (but not superposition) by using its local and global similarity measures between each possible pair of amino each residues in a single round of dynamic programing with secondary structure-specific gap penalties. This penalises the introduction of "gaps" (meaning non-aligned residues) in alpha helices and beta sheets, and favours the formation of gaps in loop regions. For those alignments with the best global alignment scores, Kpax uses the initial residue equivalences to calculate a 3D superposition which is then refined using a few further rounds of dynamic programming with a distance-based Gaussian similarity scoring function. In other words, Kpax makes a clear distinction between structural alignment and structural superposition. It only attempts to superpose pairs of proteins that have good alignment scores (user definable threshold). It does not waste time trying to superpose pairs of structures that do not give good initial alignments.

For rigid structural alignments, Kpax tends to call fewer aligned residues than other structure alignment algorithms, but the superposed structures often give lower root mean squared deviations (RMSDs) between the aligned backbone alpha carbon atoms.

For flexible alignments, Kpax applies its rigid alignment algorithm recursively to the dynamic programming problem in order to divide the task into multiple sub-problems. The resulting flexible structure alignment typically consists of two or more tightly aligned rigid segments.

For multiple structure alignments, Kpax treats one of the structures as a rigid pivot structure, and it then rigidly or flexibly aligns the remaining structures in a pile-up manner onto the pivot structure. By default, each structure is treated in turn as the rigid pivot, and the alignment that gives the best overall alignment score is retained as the solution. It is also possible to use a user-defined pivot structure.

What Else Can it Do?

As well as being very fast, Kpax writes complete PDB files of the calculated pair-wise superpositions, and information about each structural alignment and superposition in several formats. It also writes command scripts for the Hex and VMD programs in order to let the user easily visualise the calculated alignments in a variety of different drawing styles. In database search mode, Kpax automatically calculates receiver-operator-characteristic (ROC) curves to measure its own performance, and it can be used easily with both the CATH and SCOP databases. It is equally easy to build and search user-defined structural databases.

A particular strength of Kpax version 5.0.0 is that it can calculate multiple flexible structure alignments. Another useful new feature is that it can calculate the quality of structure alignments produced by other structure alignment programs.

Installing Kpax


Kpax currently only works on Linux systems. This is mainly because it makes heavy use of the Linux directory structure to store its data and to organise its output. In the future, Windows and Mac versions might also become available. In order to run Kpax on Linux, you will need a relatively recent Linux distribution such as Linux Mint 17 or Ubuntu 14 . The Kpax binaries were compiled on 64-bit versions of Mint 17.2. The latest binary has been linked statically. This means it should run on other Linux distributions.

The Self-Installer

The easiest way to install Kpax is to download and run the self-installer script. Assuming you have a 64-bit system, open a command terminal and enter something like this:

sh kpax-5.0.0-x64-mint17.bin

The self-installer script will ask some questions about where you wish to install Kpax and whether you will let the script modify one of your shell start-up scripts in order to run the program from the command line. The installer will normally define an environment variable called KPAX_ROOT, and it will add ${KPAX_ROOT}/bin to your command path. It will also run a test script located in the ${KPAX_ROOT}/test subdirectory. If you do not trust the self-installer script, you can extract and inspect the installation tar file as follows:

sh kpax-5.0.0-x64-mint17.bin --noexec cd kpax-dist-5.0.0 gunzip kpax-5.0.0-x64.tgz tar vtf kpax-5.0.0-x64.tar

You can then create your own installation directory, define the KPAX_ROOT variable, and install the installation files yourself. Please see the file ${KPAX_ROOT}/doc/README for further details.

Running Kpax

Superposing Structures

Kpax can compare and superpose multiple protein structures in a single run. Each structure must be given as a separate PDB file. Normally, the first PDB file is treated as the "query" structure, and any subsequent PDB files are taken as the "target" or "database" files which will be compared to the query structure. The following example compares one query against three "target" structures from the Kpax test directory:

cd $KPAX_ROOT/test kpax d1bhga1.ent d1cs6a3.ent 1qb5D00 3ullA00

Here is the typical output:

Kpax 5.0.0 starting at Mon Sep 21 15:58:11 2015 on host hardy. Creating RESULTS directory: ./kpax_results/ Using LOG file: ./kpax_results/kpax.log *Warning* Chain break A:S-51 <--> A:D-61 in file 3ullA00 Creating RESULTS directory: ./kpax_results/d1bhga1 Done 3 alignments for 3 targets in 0.0066 seconds (458/s). Done 3 superpositions in 0.0453 seconds (66/s). Writing QUERY pdb file: ./kpax_results/d1bhga1/d1bhga1_query.pdb Writing TARGET pdb file: ./kpax_results/d1bhga1/d1cs6a3_d1bhga1.pdb Writing TARGET pdb file: ./kpax_results/d1bhga1/3ullA00_d1bhga1.pdb Writing TARGET pdb file: ./kpax_results/d1bhga1/1qb5D00_d1bhga1.pdb ===================================================================================================================== Top 3 matches for d1bhga1 (length 103) [] -> [] (3/3) ROC AUC = 1.000000 ===================================================================================================================== Rank K-Score G-Score J-Score H-Score M-Score T-Score RMSD N/* P/$ I/@ I/% Len Seg TP Match [Family] ===== ======= ======= ======= ======= ======= ======= ==== === === === === === === == ================= 1 39.99 43.27 0.4130 0.4469 0.5538 0.6255 3.07 79 82 10 12.7 91 1 +1 d1cs6a3[] 2 36.62 25.31 0.3505 0.2422 0.2812 0.3243 3.15 45 81 3 6.7 106 1 +1 3ullA00[] 3 29.33 10.78 0.2904 0.1068 0.1381 0.1894 4.47 33 68 1 3.0 99 1 +1 1qb5D00[] ===================================================================================================================== Mean 35.31 26.45 0.3513 0.2653 0.3244 0.3797 3.56 52 77 4 7.5 98 1 3 ===================================================================================================================== Done 1 queries in 0.11 seconds (0.00 minutes). Peak memory allocation: 12 Mb. Total memory on exit: 0 Mb. Kpax finished in a total of 0.18 seconds (0.00 minutes).

File Naming Conventions

The above output shows that Kpax has created a directory called $KPAX_ROOT/test/kpax_results/d1bhga1/, in which it has written the original query structure (now named d1bhga1_query.pdb) and the three superposed target structures, Except for the first query structure file, the naming convention of the output PDB files is always <target_name>_<query_name>.pdb, which is intended to convey that the file contains the superposed target structure in the coordinate frame of the query.

This file and directory naming convention might seem verbose at first, but it becomes very useful when dealing with multiple target or database structures, and even multiple query structures because it avoids filling up your current working directory with large numbers of results files. The only file that is written to the current directory is a file called kpax.log, which contains a copy of all of the text that is printed to the terminal.

Of course, if you want to compare only two structures, you would just supply two PDB file names on the command line. If you want to suppress writing results to sub-sub-directories, you can give the "-nosubdirs" option. For example,

kpax -nosubdirs bhga1.ent d1cs6a3.ent

will run a single pair-wise comparison which will write all results files to the ./kpax_results directory. If you really want the output files to appear in your current working directory (not recommended) you can set the KPAX_RESULTS environment variable to "."

The Printed Results Table

In addition to the various output files (described in more detail below), Kpax writes a line of information for each compared target structure, on order of similarity to the query structure. From left to right, this information includes: the alignment "K-score" (calculated before any superposition); the "G-score" (calculated after the superposition); the "J-score" (normalised K-score); the "H-score" (normalised G-score); the "M-score" (the Kpax multiple alignment quality score) the "T-score" (the TM-score of the alignment, as defined by the TM-Align program); When two identical structures are aligned, the K and G scores will be numerically equal to the number of residues in each structure, and the J, H and M scores will be unity (1.0 = a perfect match).

The remaining values have the following meaning. M is the number of matched residues calculated after the Gaussian superposition (this corresponds to the number of aligned residues of other alignment programs). N is the number of initial residue equivalences calculate in the first alignment (no superposition) using the K-score. RMSD is the root means squared deviation of the superposition after Gaussian refinement (this corresponds to the superposition RMSD of other alignment programs). I is the number of residue identities found in the final Gaussian alignment (by default, residue type is not used in the alignment scoring function). Len is the length (number of residues) of the target structure.

The final two values are useful when searching CATH or SCOP databases. The last column shows the CATH or SCOP classification code of the matching database structure. If the calculation does not involve a CATH or SCOP database, zeros are shown for the classification code, as show above. On the other hand, assuming that the query structure exists in CATH or SCOP, the TP value shows whether the retrieved structure belongs to the same CATH or SCOP family (+1 means yes, the match is a "true positive"; 0 means no, the match is a "false positive").

Sorting the Output

By default, the results table is sorted according to the H-score (normalised 3D Gaussian superposition score). However, you can specify that the results table should be sorted on any of the numerical columns using the "-sort" option. For example, to order the results by RMSD, you would write

kpax -sort=R

You could equally specify K, J, G, H, T, N, M, A, or I, (all case-insensitive) as the sort codes.

Flexible Structure Alignments

By default, Kpax calculates rigid structure alignments (i.e. "-rigid" is the default). To calculate flexible alignments, just add the "-flex" command-line option:

kpax -flex d1bhga1.ent d1cs6a3.ent 3ullA00

This will cause Kpax to treat the first structure (d1bhga1.ent) as the rigid "query" onto which it will flexibly superpose the two following target structures using two structural superposition runs.

Multiple Structure Alignments

By default, Kpax calculates pair-wise alignments. To calculate multiple alignments, just specify "-multi" on the command line. For example, the following calculates a multiple alignment of three structures:

kpax -multi d1bhga1.ent d1cs6a3.ent 3ullA00

By default, Kpax will treat each structure in turn as the pivot structure, and it will return the alignment that gives the best overall M-Score (see below). To force Kpax to use the first structure as the pivot, use the "-nopivot" option:

kpax -multi -nopivot d1bhga1.ent d1cs6a3.ent 3ullA00

Kpax also supports multiple alignments of flexible structures, where the pivot structure is kept rigid and the other structures are flexibly aligned onto it before building the final multiple alignment. For example:

kpax -flex -multi -nopivot d1bhga1.ent d1cs6a3.ent 3ullA00

Here is text the output from Kpax 5.0.0 for the above example:

Kpax 5.0.0 starting at Wed Sep 16 16:28:56 2015 on host hardy. Creating RESULTS directory: ./kpax_results/ Using LOG file: ./kpax_results/kpax.log *Warning* Chain break A:S-51 <--> A:D-61 in file 3ullA00 Setting CORE threshold = 3 rows (including pivot) Induced alignment: rows=3, cols=133, T=300, L=106, N=103, C=65, M=0.6618, P=1.8864 Final alignment: rows=3, cols=136, T=300, L=106, N=103, C=65, M=0.6647, P=1.8946 MSA pivot d1bhga1 : M=0.6647, C=65 (keeping) Creating RESULTS directory: ./kpax_results/d1bhga1 Writing MSA results pivot =647, C=65to n of "gaps" (meaning non-oERps"/1qb5D00_d1bhga1.pdb = pivott quality score) t1bhga1/d1cs6a3_d1bhga1.pdb Writhe thrlity _1bhg.kmsa pivot file.lity score) t1bhga1/d1cs6a3_d1bhga1.pdb Writhe thrlity _1bhg.tes a pivotignmlity score) t1bhga1/d1cs6a3_d1bhga1.pdb Writhe thrlity _1bhg.pir1 secorison wh "gap =646 seconds (458/s). D880).

. Bee b =64al qf "gaps" (meani,ng) Creating RE1bhga1 : M================================================================================================= Done 1 queries in 0 (meaninnnnnnnnnnnnnnnnnnA(nnn226) -------TYIDDI--T---VTTSVEQ-----DSGLVNYQISVKGS-NLFKLEVRLLD-A-ENKVVANGTGTQGQLKVP--- hresholnnnnnnnnnnnnnnnnnnA(nnn 10) -LERSLNRVHLLGRVG---QDPVLRQVEGKNPVTIFSLA nnnnnnnnnnnnnnnnnnA(nnn209) ---RQYAPSIKAK--FPADTYALTG name='tQMVTLECFAFGNPV-PQIKWRKLDGSQTSKWLS-S-EPLLH-I-Q--- ---------------------------------------------------------- --> nnnnnnnnnnnnnnnnnnA(nnn269) ---------NVDF-------EDE-GTYEC---EAENI----KGRDTY-QG-RIIIHA ---------------------------------------------------------- --> 7

Multiple Struct7ut

ultiple Alignments s well as being very fascation/p> id aligns receiver-op1.0 = and it canAlignments version 5.ise aliculated nAlignment aflexible structuresnt scores . Thibove. grams, provided theograms.

Kpax t;target_nbove. positionrmalised 3Df "gapperposed ructtwo following tare re fe output from K To caltext> ase strure d "the thrlity _1bhg.tes a =codeplain>Kpax 5.0.0 starting at>ithe thr1bhgaA ------TYIDDI--T---VTTSVEQ-----DSGLVNYQISVKGS-NLFKLEVRLLD-A-ENKVVANGTGTQGQLKVP--- ------GVSLWWPYLMHERPAYLYSLEV---QLTAQTSLGPVSDFY-TL-PVGIRT >iting TA1bhgaA LERSLNRVHLLGRVG---QDPVLRQVEGKNPVTIFSLA iting TA1bhgaA --RQYAPSIKAK--FPADTYALTG name='tQMVTLECFAFGNPV-PQIKWRKLDGSQTSKWLS-S-EPLLH-I-Q--- name='tNVDF-------EDE-GTYEC---EAENI----KGRDTY-QG-RIIIHAThis will cause Kpax toure asstructurlated esame&g tlts/d1n tost direcned in a own ption=codeplain>kpax -multi -nopivot d1bhga1lated the thrlity _1bhg.tes aThis will cause Kpax to treat thcture ates a multiple p class=odeplain>kpax -multi -nopivot d Wed Sep 16 16:28:56 2015 on host har31:04reating RESULTS directory: ./kpax_results/ Using LOG file: ./kpax_results/kpax.log *Warning* Chain break Loado the same CAithe thr1bhgaost dn the comithe thr1bhg=====Loado the same CAiting TA1bhgaost dn the comiting TA1bhg=====Loado the same CAiting TA1bhgaost dn the comiting TA1bhg======Makivot =64ts outhe naming secoy aligerp136the "-so... -bit systn th, the K and G scoorydllowing tare... ctooss-checkt =64r datn th, the K an. (meaniA1bhgaaaaaaaaZEROles. sheeiting TA1bhgaaaaaaaaZEROles. sheeiting TA1bhgaaaaaaaaZEROles. shee0 Mb. es. she =aZERO d = 3 rows (incass=r CAT= 2) Induced alignment: r Asioysivot =64(3 just supply 136the "-so)... cNo.ase EMPTYthe "-sorrrrr(N==0)K To0 /136t=). D%)cNo.ase SINGLEthe "-sorrrr(N==1)K To374(37/136t=)27.2%)cNo.ase PARTIALthe "-sorrr(N>=2)K To99 (99/136t=)72.8%);nt pro=)2.17cNo.ase FULLthe "-sorrrrrr(N==3)K ToRESU65/136t=)47.8%);nt pro=)1.98 cNo.ase s (ince "-sorrrrrr(N>=2)K To99 (99/136t=)72.8%);nt pro=)2.17ccNo.ase IDENTITIES4r datnIVOT just suppon 74(17/103o=)16.5%)ce0 Mb. Rment pro(T)ivot00=Longee b> A:D- (L)ivo106==MGaussianALLthe "-so)T==). reat2 pivott quality score) t1bhga1/d1cs6a3_d1bhga1.pdb WA1bhg_msa.kmsa pivot file.lity score) t1bhga1/d1cs6a3_d1bhga1.pdb WA1bhg_msa.tes a pivotignmlity score) t1bhga1/d1cs6a3_d1bhga1.pdb WA1bhg_msa.pir1 pivotigVOT Aithe thr1bhgat on the coa1/d1cs6a3_d1bhga1.pdb WA1bhg_nt: b file: ./kpax_resuliting TA1bhgat on the coa1/d1cs6a3_d1bhgaiting TA1bhg_ithe thr1bhg=====e: ./kpax_resuliting TA1bhgat on the coa1/ TA1bhg_ithe thr1bhg=====e: ./kpa =64it matice coa1/d1cs6a3_d1bhga1.pdb WA1bhg.kit n: 12 Mb. Total memory on xit: 0 Mb. Kpax finished in a total of 0.18 seconds (0.00 minut0s).


    Superposing S3ructures a User Database
  1. ned in sraere siy agaefulp devH oro orded resiation d structuraon runs. e futpuis equally wing calculathe resuldefined ructures. which"ga1/" PD-line ofrectuery agase it avoids fillthe fion. For ithe Hex a=codeplain>kpax -multi -nopivot d1bhga1defin=

    1qb5D00 3ullA00

    Here is the typical as being very ur own instalructura calcuctures.unds of e Hex and"http:=codeplain>kpax -multi -nopivot d1bhga1db=

    1qb5D00 This will cause Kpax to treat thhe ot

    <(e inOT}/tep clas)=codeplain>Kpax 5.0.0 starting at Wed Sep 16 16:28:56 2015 on host h6:05:45reating RESULTS direc_resulpax_results/ Using LOG file: ./kpax_results/kpax.log *Warning* Chain break _resulDATABASEults/ Using /h / to/a>ed/ < *Warncuctureshga1/_*War/c_resulpax_results/ Using LOG file: ./kpants for 3 tar4ets in 0.0066 se4onds (458/s). Done 3 superpo611ions in 0.4453 seconds (66/s). Wr54ing QUERY p74 file: ./kpax_results/d1bhga1/d1bhga1_query.pdb Writing TARGET pdb file: ./kpax_results/d1bhga1/d1cs6a3_d1bhga1.pdb Writg TARGRGET pdb file: ./kpax_results/d1bhga1/1qb5D00_d1bhga1.pdb =iting TARGET pdb file: ./kpax_results/d1bhga1/3ullA00_d1bhga1.pdb Writing TARGET pdb file: ./kpax_results/d1bhga1/1qb5D00_d1bhga1.pdb ===================================================================================================================== Top 3e 1 queries in 0 ga1 4length 103) [] -> [] (3/3) ROC AUC = 1.000002/4==================================================================================================================== Rank K-Score G-Score J-Score H-Score M-Score T-Score RMSD N/* P/$ I/@ I/% Len Seg TP Match [Family] ===== ======= ======G-Sco==== ======= ======= ======= ==== === === === === === === == ================= 1 39.99 43.27 0.413.0..olnnn.0..olnn======nn======nn======nn======nn). Donnn.0.nnn.0.nnn.0.n100.onnn.0.nnn0] 2 0.0.062 25.31 0.3505 0.4469 0.5538 0.6255 3.07 79 82 10 12.7 91 1 +1 d1cs6a3[] 2 36.62 25.31 0.nnn0 0.2422 0.2812 0.3243 3.15 45 81 3 6.7 106 1 +1 3ullA00[] re is the [2.4 0.145.31 04422 0.1068 0.1381 0.1894 4.47 33 68 1 3.0 99 1 +1 1qb5D00[] = re ========2.4 0.5================================================================================================== Mean 35.31 26.45 0.351352.23.7 1.507 13555 39255 9356 534 1 2.63.0 6106 .2904 1qb 8200[] = 2================================================================================================ Mean 35.31 26. st gives low)gd radiussidue in doe fo. Kpax tNoto (ding up y/a> exnment tnment pror dataisresulatoms. ,esidues in each strucCA aild bsidusse shnsificdues ofitructure (now nly, the first PDi>!-- -----------3-------------------------------------------------------------------- -->

    File Naming C3nvention>

  2. ost ds filltpt, riptsr example, tton. For exarmlir exampl(up yw dat <

    ucturesscup y

    =c $KPAX_ROOT/test kpax d1bhga1.ent rm -rf1/1qb5D0cuctureshga1/_*War/ci>!-- -----------3-2------------------------------------------------------------------ -->

    The Printed R3sults Tase

  3. ructures.tput. before , aen calterOOT/i idrlaxat is wot recew dat/a> ludes: oicora of the nt ti

    nt vaDATABASEu really want theur own iobarge ttrucCA doe foe run. /v3.4.0/CathDoe foPdb.S35.v3_4_00-x6">idrlaxat is -bit system, open th a cCA the naming consblems.ay_ROOT/test/kpax_r/emoh/doenten>at is yfy K, J, nstalimnts nstmblems./a> on. For ex multiple a Hex a=codeplain>kpax -multi -nopivot d1bhga1defin=emoh> urce=/emoh/doenteodeplain>kpaition to the varAscripts ur sheattarge. For exc Hex asser-defined oseshcCA ned in ds to t results/d1bhdefinhdefin_emohfirst PDi>!-- -----------3-------------------------------------------------------------------- -->

    Sorting the O3tput

  4. !-- -----------3-4-------------------------------------------------------------------- -->

    Weaknesses? =====P be used e By default, Kpax calculate tihe nary ates a/a> In other w guaraxt>urrentlongs too"-nopiv pivotx calcu at is bate a. For exes. s no,ion of

    ror datrellatio. CA ((+1 mea)databasethe Kpax mu pivot <5.ise alicues receiver-operator-chinux teristic (ROC) curves tore its the Kpan be used e own p target ET pdbtrust the sres because it avoi G sB file Each structuren>at is r you will lealigalignment ofEacggreg-chi====own p gnmsumm. For exledltipuw named >at is ====own psasily with ignment or exledltipuw n PDggreg-chi G aPund. r excwn p (AUts the Kpaogram from ttial4ch struct using tmance, agramlow). n be used eithat is wrm K Tocation/abase" AUt" thitructu -----------3-5_6--------3-------------------------------------------------------------------- --> K4ax/a>


    Superposing S4ructures o(meaworkinstrun th By default, the results table is s target owing tare ,line of information insta Normally, the in severe targe. Fos printed t aost direcre (now ninn more dtion does nn rewilaled, the K andh both directory ch structres -mati nly, the ounds o ion doe"-unifi "to ordere folloodeplax -multi -nopivot d1bhga1unifi ..

    File Naming C4nvention/i> versio

    sKattarge. For exarget" protein struthe because it avoi,o ion doee and superp If you really lealiga target It also

    The only file >at is -bit system, open tein struthewotuery agase it avoids fillodeplax -multi -nopivot d1bhga 1qb5D00 3ullA00

    The remaining values hvely recefindor ex multiple consbleor ex/1qb5D00_d1bhga1.pdb =p If you really =codeplain>kpaodeplaiulentl rting TARGET pdbtes a –tr ex strulignment prbleo file. thithe thrET pdb fiaA ----TYIDDITV---TTSVEQDSGLVNYQISVKGSNLFKLEVRLLDAENKV----VANGTGTQGQLK-VPGVSLWWPYLM HERPAYLYSLEVQLTAQTSLGPVSDFYTLPVGIRT >iting TARGET pdb fiaA RQYAPSIKAKFPADTYALTG naQMVTLECFAFGNPVPQIKWRKL naDGSQTSK-WLSSEPLLHIQNV-DFEDame='tc------EGTYECEAENI----KGRDTYQGRIIIHAThis waodeplaiulentl rting TARGET pdbk stru –ttiple aon). t maticr datSSEs orrrrr(*ulated af+ $ un.uted a): 69 + 0ivo64 #SSEseplaaaa(AivoALPHA, BivoBETA, ====COIL/LOea) #========================================================= Mean 35.31 26. ----TYIDDITV---TTSVEQDSGLVNYQISVKGSNLFKLEVRLLDAENKV----VANGTGTQGQLK-VPGVSLWWPYLM RQYAPSIKAKFPADTYALTG naQMVTLECFAFGNPVPQIKWRKL naDGSQTSK-WLSSEPLLHIQNV-DFEDame='tcaaaa**@*****aaa*****aaa**@*******@********@*@aaa*** aaa*********** @ **** aaa HERPAYLYSLEVQLTAQTSLGPVSDFYTLPVGIRT ------EGTYECEAENI----KGRDTYQGRIIIHAaodepla****@******epla***@*@*****@** wa#========================================================= Mean 35.31 26. ----cccccbbb---bbbbcccccbbbbbbbbcccccbbbbbbbbcccccc----ccbbccccccbb-bccbbccccccc bbbccbbcccccccbbbccc---cbbbbbbbbccccccbbbbbcc---ccccccb-bccccccbbbccc-cccc----'tcaaaa**@*****aaa*****aaa**@*******@********@*@aaa*** aaa*********** @ **** aaa cccccbbbbbbbbbbbbcccccbbbbbbbbbbbbb ------cccbbbbbbcc----ccbbbbbbbbbbbb odepla****@******epla***@*@*****@** wa#========================================================= Mean 35.31 26. The final twulentl rting TARGET pdbknAlis –ttiple aon). t matiche o-directin th-5.0ed es orrrrr(*ulated af+ $ un.uted a): 69 + 0ivo64 #T.utrix (T ivoT.utrix * T In): # -0.912632a 007884 -0.027094 71.614277 # -0.395763 -0.865025 8108388)72.981215 #4 423501 0.21671 950875a52.880029 #4 ======4 ======4 ======4=========#========================================================= Mean 35.31 26. # Pos.351Des core H-Score M-Scorly] II/% QbecauuuuuuuuT =#========================================================= Mean 35.31 26. plaa1plaa-1.==nn). Do==nn). Do==nn) =nn) -deplaaaaa A:209:ARG) - plaa2plaa-1.==nn). Do==nn). Do==nn) =nn) -deplaaaaa A:210:GLN) - plaa3plaa-1.==nn). Do==nn). Do==nn) =nn) -deplaaaaa A:211:TYR) - plaa4plaa-1.==nn). Do==nn). Do==nn) =nn) -deplaaaaa A:212:ALA) - plaa106 2.09nn). 1617nn). 5710.0.0] A:226:THRaaa A:213:PROaa * plaa0.0.01 97nn). 5132nn). 88 1 ] A:227:TYR) A:214:SERaa * plaa 1 31.5538a 244 1 . 81381 ] A:228:ILE) A:215:ILE) @ plaa81 31.56nn). 54211 . 81631 ] A:229:ASP) A:216:LYSaa * plaa906 2.23.7 0.57nn). 5300.0.0] A:230:ASP) A:217:ALA) * pla101 31.57nn). 403555). 73 6.7 ] A:231:ILE) A:218:LYSaa * pla1106 2.20nn). 1691nn). 53 1 +1 A:232:THRaaa A:219:PHEaa * pla1 1 + 5 88nn). D778nn). D1211 =nn)A:233:VALaaa A:220:PROaa * pla13plaa-1.==nn). Do==nn). Do==nn) =nn) -deplaaaaa A:221:ALA) - pla14plaa-1.==nn). Do==nn). Do==nn) =nn) -deplaaaaa A:222:ASP) - pla15plaa-1.==nn). Do==nn). Do==nn) =nn) -deplaaaaa A:223:THRaaa- pla10.0.017.52 aa 2 11nn). Do=71 =nn)A:234:THRaaa A:224:TYR) * pla1 1 3 D1nn). 3814pla. 12871 =nn)A:235:THRaaa A:225:ALA) * pla181 3 4555). 2192nn). D7=======nn)A:236:SERaa A:226:LEU) * pla1906 2.8=nn). 1589nn). 36 91 =nn)A:237:VALaaa A:227:THRaaa* pla2006 2.04pla. 0154pla. 5884 1 =nn)A:238:GLUaaa A:228:GLYaaa* pla21plaa-1.==nn). Do==nn). Do==nn) =nn)A:239:GLN) ) -deplaaaaa- pla22plaa-1.==nn). Do==nn). Do==nn) =nn)A:240:ASP) -deplaaaaa- pla23plaa-1.==nn). Do==nn). Do==nn) =nn)A:241:SERaa -deplaaaaa- pla24 1 5 77nn). 1646pla. 01 3 =nn)A:242:GLYaaa A:229:GLN) * ... (e inOT}/t)The remaining values have th If you really l receealiga targeor ex multiple consb(text> iniignt afrira)=codeplain>kpaodeplaiulentl rting TARGET pdbprofit –troe aon)ltpt, structural sProfitrrr–t t matic ot tcaalslb tosyare cytrucCA s yes,rrr–tx-yxc query. rooitiplota> rrr–t VMD prdowdrawter Gaussian refineural sn ouperrrrrrrrrrrrr–t VMD prdowdrawter Gaussian refineural sto rrrrrrrrr /p>

    The Printed R4sults Taso(meaworkinstruAm, apint

    iselook (d1bhga1.enthat fu e futpulculated 1btt d easpe is not us/a> classmu wouts t 40length 1btrust thur own ichangd ese/p> a < $rhunds o -he othly - t o ordedr the results by ex multiple a Hex athe varr ledup y calJGaussiaentlJ-Scer Ga structuret is pr run. Ea CA cucturesscore nstag very owing targample,nto cal t 200length 1eealigam fr a b, JGaussiaa structur:dition to the var $KPAX_ROOT/test kpa1bhga1 t=200l-he o=200l-db=emoh> rt=Ja 1qb5D00 3.. _>!-- -----------4_2_1'> Kure AliPa < $rligerpO orded

    Superposing S5ructures

ally It alspa < $rliof all of eaw alslon runs. a paces =0.5> blo t =0.on-ipld o=33.. _>!-- -------the output fr, te To cal meml by the TMrlated eIt /liipld odefi+/-ot) ment prbost diree outpunment px otle /p>;xamppaces "K To caltepositionrmalppaces ry structure1.0 =e outpu grams, prounds o c TM, agram_ROdefined nedteinscore " blo t "K To cale outpuBlo t 62 yens.acs (i structure1.0 =btrust thKPAX_ROorce calBio thec. emaining values have nax 5a Hex agnclassificg QUERacause it avo-e commicagaptn .. _>!-- -------the output fr, rho"K To calureican pen (d0.1),gen in sever"- "K To calgaptax codeCATc pen (de cl);n sever"-he ", b "score " gaHex"get" ned n kpax -multi -nopivot d1bhga1leeing-cycle=1 -co(m cu=8 .. _>!-- -------the output fr, leein"s a "faues. taiguribefragm> pair-lences con runs. a eei);n sever"-cycle"K To caleach strucPAX_t-he sups,fita> h-5.0ed e cluding piture asthe var nAlis rucCA ar dat-5.0ed esigled $rh shnsif To cuding piaCOP, nmentdaost direcg very owing taiordehe olve a CA/tba2_3> !-- -------g S5ruc------------------------------------------------------------------- -->

File Naming C5nventionMted O ordedkpax -multi -nopivot d1bhga1hhlpuThe remai-----5nvenemai-----5he fin--------------------------------------------------------------- --> K6axRe she a2_1'> Kpaave multiple Kpcor alees where the pive builctures and a> ally. Thisubmireca=codeplaa> ulentl codeplaa> i>C olve a r Datab Inh in , H,tureees whernedteinre the pive builcture .codeplaa> D.W. Rto/a>e (2016).codeplaa> buleaodepl> bp_1'> Kpaave a i>Kpcorn belowsdirecre (now n/a> nAlignmentograms.

algre thm=codeplaa> ulentl n/a>FX_tsProteinrh4>

s and score). OIt lapt>ultipleaftcodeplaa> Btoms. Pe orde Fragm> S structur .codeplaa> D.W. Rto/a>e, A.W. Ghoorah, L. Mavri-5., V. Venka xam (2012) thec..ox djourow /bts618?ijkey=EUTXiZpeziFL te&">Bio thec.stal,codeplaa> 28(24) 3274-3281,iatag very to/a>ed/ aus m> acauMd $res axat is u/l='codeplaa> buleaodepl> bp_1' remai-----6he fin--------------------------------------------------------------- --> wrm K Toam error ineodeplain- -multi wS-51 <> wrm K Toa wS-51 < ineodeplain- -multi succuss> wrm K Toa succuss ineodeplain- -multi > wrm K Toam the al ineodepla2_6--- bpax >

  • div> ---------d> ---------r_1'> ---d> -----r_1---ng t> -----ng t-multi foo $rhbam fr=0 c llspac1 <=5ar dth=600pxhe fin--r_n-- cols=1=c TMeraddition" title="ave Intiatrelucturac TM, ">Intia4>Multiple     dition" title="Labam iral mrrrgern ux t t em theq pxxcep 1eepllic a CAd">LORIA4>Multiple     dition title="ave Caps team">Caps Team4>Multiple     dition mav- title="UnristaioĆ©rn uLmrrrgee">Unristaioérn uLmrrrgee4>Mution - ---d> ---r_ 1---ng t> - -