Monte v2.02 User Manual by T. Kevin Hitchens
Contents
1. to a particular data type The next section 1s used to adjust the relative scaling of different data types The only exception is the relative NOE scale this may be changed in the simulated annealing parameters see Simulated Annealing parameter section Entering a scaling factor of zero will turn off a scoring of that particular data type In the example above data from the N specific label file will be multiplied by 2 during the Monte run The entry NNn and NNc are for the N terminal Nitrogen and C terminal Nitrogen chemical shift matching see Chemical Shift Definitions section O PO A 5 44 Tolerence for Chemical Shift Matching Monte scores matching between y f Tolerances for chemical shift matching sigma ppm for Gaussian complementary chemical shift pairs 1 e a a a a ae E HA he CA of resi with the CA of ERE ES A Baan neon aa E a A loto foos foio fogo foio foro foto foto difference between the chemical shifts to a Gaussian distribution function An exact match will be scored as 100 multiplied by the weighting factor The N entry is used for matching J coupled shift obtained from the H N COCA NH type experiment as well as amide NOE matching and matching specific label shift in conjunction with the H atom entry The chemical shift matching is truncated at 1 00 ppm for all chemical shift matching In the example in Table 1 a score of O is given to any pair of shifts that have a difference in chemical2. 0 04 therefore the CA shifts between residue 21 and 22 are highlighted in the medium yellow shade indicating that the difference in the shifts between 2 3 times the chemical shift tolerance HIS 320 20 173 FOAD S514 52 74 1 3 73 178 06 lt 1 gt 18 19 21 ALA 1140 21 16942 12402 592 Size 1820 1 33 Sir 19 19 20 22 235 24 GLY 20 22 841 104 69 46 92 5582 0 00 178 17 lt 0 gt 21 23 24 25 GLU 125 0 23 769 124 37 5845 46 94 179 67 176 35 lt 1 gt 21 22 24 25 26 wi Residue numbers and residue type names that are shown in bold face type indicate that this assignment 1s consistent with the highest 50 scores solutions for the last annealing step Entries that are not bold face type indicated that the peak assignment for this residue changed over the 50 highest scores The bold type residue is an indication of a possible correct assignment One must further verify the goodness of this assignment by evaluating the goodness of chemical shift matching NOE matching residue type prediction and specific labeled sample matching As well as the consistency of this assignment over the course of several Monte runs Note the inclusion of an assignment in the 50 highest scored solutions is dependent upon the amount and quality of the data input relative scaling factory and annealing schedule used Care should be taken when analyzing these results Following the chemical shift is a number in brackets lt gt this number indicated the number
3. 1 It may be advantageous to start out a Monte run with a scaling factor of zero As the melting temperature of the system is reached slowly increase the NOE scale to desired value 8 16 Repulsive Terms As Monte assembles linked spin systems it may be 100 advantageous to penalize linked assignments that 75l have chemical shift differences much larger than the E desired tolerance set in the parameter file In this del H i case one may add a repulsive term to the chemical Score 25 shift matching calculation Enter the desired o Ed repulsive term in the field for the appropriate atom b A type The baseline of the Gaussian matching oem function will be lowered by this value For example 50 o _____ if a repulsive term is set to 50 The matching score 40 45 50 55 60 65 70 for an ideal match will be given a score of 100 a chemical shift difference greater than ca 3 times the tolerance will be given a score of 50 as shown in the figure to the left In all cases however a missing chemical shift i e and entry of 0 0 will not penalize a solution Chemical Shift ppm 8 2 Saving the Parameter File After all the desired parameter values are set one may save the parameter file by clicking Save Parameter File The software will save the parameter file and check to see if the necessary files are available the gui will warn the user if there are any inconsistencies between the files marked as
4. The sequence file provides a primary sequence of the protein being studied If a pdb file is available the sequence file 1s not necessary The sequence file is identified by the extension seq and must be in the assignment sub directory The sequence file contains in primary sequence order the three letter codes capital letters for each amino acid separated by one or more spaces and or carriage returns Each string of text may not exceed 100 characters 25 Amino Acids if separated by a single space MET ALA GLY PRO BS GLY LS LEU The sequence file in conjunction with the secondary structure file is used to create a pseudo pdb file for NOE matching For a coil or sheet secondary structures a linear peptide is created for residues identified as helical segments an a helical peptide will be created This pdb file is written to temp pdb 4 5 PDB File The pdb file provides both the primary sequence of the protein being studied as well as amide proton coordinates for matching NOE s Monte will add amide protons to a crystal structure The pdb file is identified by the extension pdb If a pdb file is used as input a seq file is not used or needed Note The pdb file should end in TER 4 6 Secondary Structure File 6 The secondary structure file is identified by the extension ss The secondary structure file provides information on the secondary structure of peptide segments This information is used for
5. available and files present in the experimental sub directory Open Es 1D table Open par file save par file a tart Kill Quit i 16 9 Data Output files Monte writes several forms of output during the simulated annealing run as well as at the end of the run These output files are useful in determining the quality of an assignment solution and help the user make decisions concerning the annealing schedule and other parameters The log file 1s also useful to trouble shoot problems that may arise with files that may have a format which is incompatible with Monte 9 1 Log file Monte creates a log file in the experimental sub directory The amount of information written to this file 1s controlled by the log level parameter log level 0 Nothing is logged log level 1 Only the filenames of input data and standard output are logged log level 2 In addition to log level 1 most data read in from input files is written to the log file This includes chemical shift database pdb file specific label files etc log level 3 In addition to log level 2 all internally calculated data is logged 9 2 Standard Output During the simulated annealing Monte with write text to the unix shell indicating the progress of the annealing schedule An example 1s shown below anneal max max NOE sched temp neye nS uC n rej score score swap scale 2 140 0 10001 2112 12489 20 9 LL INES 2 Oy aA CO Ca Ha Cb Hb Cg Nn Nc 27145 2436 430
6. of NOE shifts that were input for that spin system The following numbers residue numbers that are predicted to have NOE s based on the pdb file and the distance cut off value A number in bold face type indicates that the predicted NOE is satisfied by an NOE chemical shift that was provided The peak number in the top frame is a hyper link that will change the data displayed in the lower frame The lower fame indicated the residue type probabilities calculated by Monte for the i and 1 1 residues for that spin system The greatest value 1s indicated in bold face type In the example here one can see that peak 2 0 has an 86 probability of being a glycine based upon the N and CA shifts note than no CO shift is input for this residue type This window also reveals that the inter residue CA and CO shifts alone do not define the residue type of the preceding residue of this spin system very well res ALA VAL LEU THR SER ILE GLY ARG ASN GLU GLN ASP TRP TYR PHE LYS CYS MET HIS PRO This information is helpful to determine if a chemical shift offset should be added or subtracted to the input shifts in calculating residue type information see Parameter File section 19 9 5 Assignment Summary Postscript A graphical summary in postscript form of the final assignment solution is provided in the main directory as the file filename_out ps An example output file is shown below left side and an enlarged view of the image is sh
7. what the names of the applications are This is done by editing two lines at the beginning of the tcl tk script fm tcl e set browser netscape e set psviewer xpsview You should replace netscape and xpsview with the names of the browser and postcript viewer on your system 10 2 Running Monte 1 The most important thing to remember is to save the parameter file before starting a run Monte always reads the parameter file it does not directly obtain the information from the tcl tk window 2 The initial annealing temperature should be high enough so that most proposed swaps are accepted 3 The final annealing temperature should be low enough so that the score has stabilized The annealing schedule should be slow enough to maintain equilibrium Try doubling the number of attempted swaps swap_lim at each stage of annealing and check whether the final score improves 5 Inspect the html output If all the chemical shift matches are within tolerance i e no yellow orange or red blocks you can afford to tighten the tolerances for chemical shift matching 6 View the correlation plot corr ps If long stretches of connected chemical shifts are swapped between different residues you may want to increase the scaling factors for residue type RT i RTG 1 and specific labeling nsl and csl 21
8. 0 Da 3 54 8 0 0 67 8 0 0 Os OL 993 oo 98 0 34 92 7 0 0 09 7 0 0 O02 J9 92 9 05 35 50 6 0 0 69 7 0 0 0 203 99 98 83 PALAS 48 5 0 Di 0 6 0 0 0 04 99 J6 TZ Di 46 4 0 Oil 6 0 0 O05 98 94 60 0 38 44 4 0 On TZ 5 0 0 O06 08 OZ 48 AS 42 3 0 iS 5 0 0 0 07 oF 89 od 0 40 41 2 0 0 74 4 0 0 0 08 96 86 Zed 0 41 39 2 0 Us 19 4 0 0 009 99 83 19 0 42 oF 1 0 IAS 4 0 0 OLY 94 80 LS 0 43 oS 1 0 OL TT 3 0 0 O iL 03 76 8 0 44 34 1 0 0 78 3 0 0 0 12 92 72 5 0 45 Se F 0 Os tI 3 0 0 Orcs 91 68 3 0 46 30 0 0 0 80 2 0 0 0 14 03 64 1 0 47 29 0 0 isso ll 2 0 0 O Lo 88 60 1 0 48 a 0 0 DSZ 2 0 0 OMIE 86 536 0 0 49 20 0 0 Ox 63 2 0 0 Qa 85 52 0 0590 24 0 0 0 84 1 0 0 Oo 83 48 0 Oc DL 23 0 0 0 5 8 0 i 0 0 Dades el 44 0 Ds 0 22 0 0 0 86 1 0 0 0320 80 41 0 005 Al 0 0 oey 1 0 0 al 78 of 0 0 54 t9 0 0 0 88 1 0 0 22 76 34 0 006 L7 0 0 O39 1 0 0 PS 74 30 0 Doa 16 0 0 0 90 i 0 0 0 24 72 27 0 0 58 15 0 0 0 91 1 0 0 Us 20 FO 24 0 Uva 14 0 0 Oe92 0 0 0 0 26 68 22 0 060 LS 0 0 O93 0 0 0 Oise 66 19 0 O64 12 0 0 0 94 0 0 0 0 28 64 17 0 0 62 alia 0 0 00095 0 0 0 0 22 62 TS 0 0 03 Li 0 0 0 96 0 0 0 Os 30 60 L3 0 0 64 10 0 0 AAN 0 0 0 Os 3 L 58 Li 0 0 60 50 J 0 0 0 38 0 0 0 OZ 56 10 0 0 66 8 0 0 0 99 0 0 0 10 0 0 0 0 12 6 Chemical Shift Identification Table The chemical shift identification table allows the user to define the chemical shift types that are listed in each column of the cs file The N and the H entries are the parent amide nitrogen and ami
9. 4 2132 6387 0 0 ee 1 eras CNO LECALSS NA NH HA HB 6111 4308 0 0 NOE 548 The top row of this output shows the current step number of the annealing and the current temperature The value ncyc indicated the number of cycles for that temperature The values n suc and n rej indicate the number of successful swaps and number of rejected swaps respectively Score and max Score indicate the current overall score for the solution and the highest score that has been reached over the current annealing step Max swap indicated the maximum number of residues that are swapped and NOE scale indicated the current scaling factor for NOE matching The next section of text indicated the total score calculated for each type of chemical shift matching These values are a result of the internal score multiplied by the scaling factor The values reported here might be negative 1f repulsive terms are greater than zero in the annealing step parameters see Repulsive Terms The following section reports the score given for residue type matching The first two indicate the total score for matching residue type probabilities based upon chemical shifts for the 1 and 1 1 residues respectively The value for N 15 reports the total score given for matching N specific label data and likewise the 1 C 13 value reports the total score for any 1 C N type specific label data The next section in this row reports the total score for NOE matching Curre
10. REO BEER 10 5 42 Residue Type Distribution Width amp Deuterium Isotope ShifES ooooonnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnos Il VA DNV CUOMO CIONS septate saci ero A II o o E eee cere Il 344 1 Olerence for Chemical Shift MGI CHU Eee 11 DD Chemica AS MUDO SOIS tt 12 6 CHEMICAL SHIFT IDENTIFICATION TA BLE j ccccccccccccccccccccccccccccccccccccccccccccccccccccccccccecccccccoccees 13 7 MATCHING OLD ASSIGNMENTS eseseseseseseseseoccscsesesesesesesesesesecscsesesesesecesesesesecececseseseseseseseseseseososcsesesesesesesesesese 14 oS SIMULATED ANNEALING PARAMETERS die 15 S OBE an at Bo A SCHEDULE dictadas 15 So LAIN UINDCT OF VCS E A stds OE NOO E EN 15 ALAS E tii 15 SAI SWAD MM S ONO AAA RAS ARA AAA A A E 15 SALA SWO S EAS AAA AAA ANS 16 O E ALS O A 16 Onl OdRCDULS IVE T CMS ARAS AAA RA AAA A 16 382 SAVING THE PARAMETER PIELE ti dd lin dis 16 DATA COGUTPUTEITLES e ES 17 IMA A e al iiagos a pel tn dicas ocios a LES a dia lees hdr a o da od ol os en e 17 INES A A A A A A A ee 17 EOI 2 A A A A A Teen ete E E AEE A AN 18 SN A A O A A A E LTR E A nae 18 9 5 ASSIGNMENT SUMMARY POSTSCRIPT anita aiii 20 MOTO RREDA TON PO atan toa caciones 21 10 CUSTOMIZING AND RUNNING MONTE cccccscscccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccoccocces 21 TOL COUSTOMIZIN CA NA ra a rA S 21 TOD RUNNING MONTE de r Neea 21 1 Introduction Monte is a program th
11. TM Monte v2 02 User Manual by T Kevin Hitchens April 2003 Monte 2003 Carnegie Mellon University T Kevin Hitchens Jonathan A Lukin Yiping Zhan Gordon S Rule Corresponding addresses rule O andrew cmu edu hitchens O andrew cmu edu Disclamer Monte is provided as is Neither the authors nor Canegie Mellon University provide any warranty or guarantee of program function or correctness of results Individual users are responsible for the use and inferences of Monte results Table of Contents G RS Ge Gi FL DL steerer ema tte ee A E RO aL ONE TE A ee Sena MTR 1 PINT RODU EC ICON susi A A id 3 2 LICENSE AGREEMENT 0d 3 S GET EENG STARTED ose 4 DIRECTOR STRUCTURE dr OE eR RC EOE A en aa 4 4 INPUT FILES etree nee ate en arp TS a eR Ee nT ES EO ee 5 A SU A ee eee ne ee PN oe TTR ECM eC EEE Se Dm Te eRe te Perea een ne Tee e Om nT Cee ee Tete Eee arte eer 5 In o RO Eo e er E OOO ieee 5 OPO AAU RA Po UE o A o A 5 4 2 FILES AACONV ANDIAACONY DEU T orainera a da dliad 5 ASES HEN ICA SHIP DATABASE raid ad dd dd dto e do tin de 5 AS SA A A A 6 A e SN A O A A A A 6 AOS CONDAR Y S PTRUETO RE E di dadas eee 7 4J PRECIO AMINO ACIDO EES ee da dona a alte 7 A ASA A a A A A cae Te RCo 8 Ss PARAMETER DIE aii dci 8 DSO VER VIEW A O R AEO sack ousbiesieudeusacwesehtanenee 8 A o A S E S AES N E none ssa A enn madcne moses netaceuencse ae E N N EOE E 9 DORA A O 10 RON PARA a o a det e dd o di oo ein 10 DAL NOE IST CO CULO A AA
12. a da o o O da lla Flo ha a ee a ee ea ee ecg Mee Pda a BL Ja a a SE a La O ela a a Ea eg he a a a la Ja a a da a La a da a a E Meda a Ba E ed a a Fal A as RE FL a LES EQU A LEO A QU AO A QU a da da Ea da Ma La O da o ka E Meda La ka EE da En EL dy SY ae Pta a oy Qe da Ll a E Meda La Aa EA ed a LS a O ela a La OE da a a E e Ma a EE da da La Ja a E SE a La a ela o ka E ed a Aa E ed a la a ela a Eta ESE da a a EE da Mi a En eta la HL a SR peda a BL EOS a Ma a EE a ME a EE ag a La da a a da a La a da a a E Moda a Ba En da a La da La E fa a La O da o a E a a Aa E dz la LO L Note about aliphatic Proton matching Monte will match HA1 and HA1 chemical shifts without the HA2 shifts being defined If both shifts are identified as in the case of Glycine residues however Monte will find the best combination of the two shifts If one desires to match HB chemical shifts however both 13 HB1 HB2 and the complement HB1 and HB2 must be defined Monte will find the best combination between these four chemical shifts If the window shown above is closed it may be opened again by selecting the Open CS ID table button at the bottom of the Main window Open Es 1D table Open par file save par file a Lart Fall Quit 7 Matching old Assignments One of the unique features of Monte is that the program can match a chemical shift assignment solution from an previous Monte output file An example of this appr
13. are constructed as a linear peptide If an NOE distance cut off of 4 5A is selected then NOEs between adjacent residues will be generated for assignment purposes 5 42 Residue Type Distribution Width amp Deuterium Isotope Shifts Monte applies a Gaussian distribution to the mean chemical Residue type distribution width sigma in ppm shifts ig pb H co CA Cc ca aaconv_deut to determine residue i imi poi ee a type for the 1 and i 1 residues 3 6 13 2 pa Fue 4 0 of a spin system One may wish x CD shifts CH shifts to widen or narrow the width of the gaussian function To do so enter the half height peak width ppm in the box for the appropriate atom type Generally these distributions need not be adjusted Directly below the distribution width table are two buttons CD shifts and CH shifts Select the button that is appropriate for your data either protonated proteins CH shifts or perdeuterated CD shifts 5 43 Weighting Factors Monte internally scores a perfect score as 100 points This maximum applies for all Chemical shift matching scaling factors must be integers matching including for matching chemical cO CA HA CB HB CG NNn NNc HH ES shift for an atom type matching a spin for fh of pf of A E system to residue type or specific labeling or NOE matching Less than ideal matching is scored appropriately less than 100 based upon user defined parameters One may wish to give more mathematical weight
14. at applies a Monte Carlo Simulated Annealing approach to obtain and identify a unique solution for the NMR chemical shift assignment problem of N C labeled proteins The motivation behind the Monte program is to provide a general software package for chemical shift assignments of proteins independent of particular required experimental data collection that may not be well suited for large or small proteins or protonated or deuterated proteins random assignment perform swap Evaluate change in score As gt le The method starts with random assignments of the spin systems to residues in the primary sequence A spin system is defined as all of the information associated with a particular amide peak This includes inter residue connectivities NOE peaks and intra residue chemical shifts All of this information is indexed by the nitrogen and proton shifts of the amide proton The initial assignment solution is scored based upon any or all of the following criteria e Chemical shift matching of correlated spin pairs How well chemical shifts identify score O the i and 1 1 residue type for a given L 9 secondary structure E e How well amide nitrogen proton chemical accept swap a shifts match data obtained from with probability specifically labeled protein samples exp As T e How well amide NOE s match the pattern predicted from a pdb file or identified secondary structure e How consiste
15. both matching residue type probabilities as well as building a pseudo pdb file for NOE matching in the absence of an X ray or NMR structure file The default secondary structure is random coil unless the helical or sheet segments are identified If a pdb file is available then MONTE will calculate the secondary structure from the pdb file However the information in the secondary structure file will override the calculation from the pdb file The existence of a ss file prevents the calculation of secondary structure from the pdb file The format of the ss file is the number of helices followed by subsequent lines that indicate the residues at the start and the end the helical segments This is followed by the number of B sheet segments followed by the starting and ending residues for each segment In the example below there are 3 helices segments 10 15 22 40 and 42 50 there are also 2 B sheet segments residues 2 5 and 17 20 These residues will match best with spin systems that more closely match the a helical or B sheet mean chemical shifts All other residues will match the mean chemical shifts for a random coil 3 TO T9 22 40 42 50 2 Ze L 20 If there are no B sheets or no amp helices a zero must be entered followed by no segments In the following example there is no beta sheet secondary structures identified 2 10 o 22 40 O 4 7 Specific Amino Acid Labels Monte will use amide chemical shift data acquired on
16. d in a file called monte license that must reside in the same directory as Monte The license key itself is a 12 character string that should be the first and only line in the license file To insure that you are using a relatively current version of Monte the license key expires one year after its issue Registered users need not sign a new license agreement Simply drop me an email and I will send you a new key For profit organizations are welcome to use Monte gratis for a limited trial period after completing the necessary right to use agreement After this period it is necessary to negotiate a license agreement Please contact Gordon Rule for additional information 3 Getting Started Extract the Monte distribution tar file tar xvf FMv2 02 tar Directory Structure The Monte home directory FM2 contains the following four files fm exe Monte pre compiled executable code aaconv Chemical shift database for protonated shifts aaconv_deut Chemical shift database for deuterated proteins fm tcl A graphical user interface for editing the project par file Once you receive the license file you should also place this file in the main directory Within the home directory one should create subdirectories for each chemical shift assignment project These directories contain the parameter par file and all user input files In addition all output files will be placed in the subdirectory The subdirectory path and experimental filename
17. de proton chemical shifts These two shifts must be in either of the first two columns of the chemical shift data base file Only one column may be selected for one atom type listed in each row To redefine a row or colum the row or column must first be deselected The intra residue 1 chemical shifts for the parent amide nitrogen pair are identified by CO CA HA etc Inter residue shifts 1 1 1 e from the preceeding residue are identified by the same atom names followed by the superscript 1 The atom types Nn and Nc refer to the N terminal nitrogen shift and the C terminal nitrogen shift These correlations are derived from the following experiments Nn H N CACO NH Ne H N COCA NH In the following example the Chemical shift matching table is consistent with the cs input file that contains the parent amide nitrogen shift in column 1 the parent amide proton shift in column 2 the inter residue CO shift in column 3 i e from the HN CA CO experiment the inter residue CO shift in column 4 1 e HNCO experiment the intra residue CB shift in column 7 and the inter residue CB shift in column 8 If there happen to be shifts in columns 9 and above the chemical shifts will be carried over to the Monte output however they will NOT be used by the program to obtain the solution PE QUE FU Y AL ce AE QUE QU 1 LO CE a a l LJ i l LJ l l LJ Zl m i Ly El i YE AAA YA S le a MA AAA da SS o a a Pa lo
18. either a specifically N labeled protein sample or a specifically 1 C carbonyl and uniformly N labeled protein sample in conjunction with shifts obtained from a 2 D HNCO filtered HSQC In the former case the pairs of amide chemical shifts identify the residue type of the 1 amide spin system In the latter case the pairs of amide chemical shifts identify amide spin systems of residues that are adjacent i 1 to the label Our experience is that the 1 C N labeling scheme is more robust and does not suffer from transaminase dilution of the specific label Matching is based on a comparison of the amide proton and nitrogen shifts in the specific labeling file to the spin systems that have been assigned to that particular residue type The score is based on the same Gaussian distribution that 1s used to score inter residue connectivities The specific label files are identified the extensions nsl for the specific N label HSQC experiment and csl for the 1 C N 2D HNCO experiment The format of these files are the three letter code capital letters for the amino acid followed by the amide nitrogen and amide proton chemical shifts The order of these two chemical shifts MUST be consistent with the chemical shift data base file and the order defined in the par file The following example is a pairs of shifts for an alanine specific label ALA 123 4 7 44 ALA 120 5 8 54 Identification of residues containing methyl groups and ot
19. ere selected and a spin system was assigned to the same residue in all of the cycles then a would appear in the Ist column Entries that do not have a indicate that two or more different spin systems have been assigned to that residue 6699 Warning The residues is an indication of a possible correct assignment One must further verify the goodness of this assignment by evaluating the goodness of chemical shift matching NOE matching residue type prediction and specific labeled sample matching As well as the consistency of this assignment over the course of several Monte runs Note the inclusion of an assignment in the 50 highest scored solutions is dependent upon the amount and quality of the data input relative scaling factory and annealing schedule used Care should be taken when analyzing these results also see HTML section Additional note if one chooses to match an old Monte output file this symbol is used to indicate a confident assignment that will be matched to the new solutions Before doing so one may wish to remove some symbols from the old output file The next column indicated the three letter code for the residue This is followed by the spin system number any assignment that is found in the comment line of the cs file followed by a list of the chemical shifts in the order of the cs file Monte uses an additional swap area that does not add any score to the s
20. freeware bin wish8 0 To do this rm usr bin tclsh and usr bin wish and then make the links as follows e In s usr freeware bin wish8 0 usr bin wish e In s usr freeware bin tclsh8 0 usr bin tclsh Running and Customizing Monte Helpful hints on how to customize Monte and define run parameters are included in section 10 of this manual 4 Input files 4 1 Overview Minimum Requirements aaconv and aaconv_deut provided Parameter file filename par provided user modified Chemical shift data base filename cs Sequence File filename seq License file monte license Optional Input pdb file filename pdb Secondary structure information filename ss Specific label data DN Specific Labels filename nsl 1 C PN filename csl NOESY data Single Amide 4D NH NH NOESY filename 4dnoe or a pair of Amide 3D H NH NOESY filename 3dh_noe Amide 3D N NH NOESY filename 3dn_noe Chemical shift assignment output from a related solution 4 2 Files aaconv and aaconv_deut These are formatted text files that must reside in the HOME directory of the Monte program The file contains the mean chemical shifts for CO CA CB CG and N nuclear spins found for secondary structures of random coil o helix and B sheet for each of the 20 common amino acids aaconv or the same information corrected for deuterium isotope effects aaconv_deut The correction for the deuterium isotope effect is selected in the Tcl tk interface The da
21. fter the number of swapping cycles is reached the temperature of the system is lowered by tstep until the temperature is less than or equal to tfin Following which the next annealing step is started At the end of each annealing step in addition to the initial random assignment a text output file is written to the experiment sub directory The output file extension is incremented as the annealing step in incremented see Output Files section Note on annealing temperature Monte is designed to provide a general method for chemical shift assignments Different chemical shift assignment problems will result in higher or lower overall scores as well as variable magnitude of score differences while comparing swapped blocks of assignments These score difference are a function of the amount of data and the relative weighting factors used for the Monte run With this in mind the melting temperature or transition between the temperature at which all swaps are accepted and better scores a preferentially accepted will vary The annealing schedule should be adjusted accordingly Monte provides some scoring and swapping data during simulated annealing process to help the user adjust the annealing schedule 8 13 swap_lim gamma The parameter ncyclim defines the number of cycles that should be completed before the temperature is lowered Since lower temperatures result in a fewer number of accepted swaps it is useful to increase the number of cycles pe
22. her residue specific experiments Additional information on the potential residue type of spin systems can also be incorporated into Monte If the residue type is uncertain e g for methyl containing residues then all possible residues should be indicated For example if the an amide peak at 120 2 and 7 3 ppm was observed in a H CC CO NH TOCSY spectrum of a methyl protonated protein then the following should be include in the csl file ALA T202 Tes VAG 1202 Po LEU TAU des EEE UZ eZ Teo 4 8 NOESY data Monte may use input from either a 4D Amide Amide NOESY or two 3D NOESY spectra an H NH HSQC NOESY and an N NH HSQC NOESY The NOESY cross peaks are matched with those which are predicted from the pdb file or secondary structure within the NOE distance cut off see Parameter file The 4D NH NH NOE file is identified by the extension 4dnoe Each set of NOE data must be assigned to a parent amide peak The file format contains the peak identification number followed by the parent amide nitrogen and amide proton chemical shifts The order of these shifts and subsequent NOE pairs must be consistent with the chemical shift data base file and the definition in the par file PARENT NOE 1 NOE 2 NOE 3 TODO Aaa daa UL Be Gans PAO Pazza od 9 00 The 3D H NH and N NH HSQC NOESY files are identified by either the file extension 3dh_noe and 3dn_noe respectively Monte will run with only the 3D H NH NOESY howeve
23. ignment solution filename_out ps A postscript figure that summarizes the assignment data solutions filename_nn soln The best assignment solutions for each cycle of the Monte run solutions corr ps A postscript correlation plot The GUI can be used to select whether the HTML or postscript output is desired If the Output postscript output is selected then you can W HTML select how many residues to print line and a W Postscript scaling factor for the final output For large os proteins e g gt 200 residues it may be Residue width 60 necessary to reduce the scaling factor to keep scaling factor 1 0 the output on one page The NOE output on W Display NOE s the postscript plot can be turned off if desired Additional detail regarding the output files can be found in section 8 5 4 Run Parameters The parameters that control the scoring are defined in the next section of the top portion of the main window You will probably have to scroll down in order to see these windows 5 41 NOE distance cut off In the next field enter the desired distance cut off A for NOE matching Monte will match NOE chemical shifts that are consistent with the current assignment solution and are within the distance cut off radius of the parent amide proton in the pdb structure file If the pdb file is not available Monte will create a pseudo pdb file temp pdb In this Distance cutoff A 5 5 10 case random coil or B sheet defined residues
24. is provided to the parameter par file for a given assignment problem All input files must share the same root filename and each individual file is identified by a distinct file extension see below Examples In addition to the above files two sample project directories called rho and rho_dna are included with the distribution The rho directory contains sample files of all of the data types that Monte can utilize It also includes a parameter file par that contains all of the parameters that control Monte Although this is a plain text file it is highly recommended that one does not directly edit this file Changes to program parameters should only be made through the tcl tk Graphical User Interface fm tcl The rho_dna directory is provided as an example of how to use known chemical shifts of an unliganded form of the protein to aid in the assignment of a liganded form in this case a protein DNA complex The file that contains the chemical shifts for the unliganded form of the protein is called rho_final out When you run the rho_dna example you will have to select this file using the tcl interface in order to get the correct directory path Running tcl tk To run the tcl GUI one should first make sure versions 8 0 4 or later of tcl and tk are installed For SGI IRIX these are available through http freeware sgi com Remember to change the symbolic links for usr bin tclsh gt usr freeware bin tclsh8 0 and usr bin wish gt usr
25. ndicated as a confident assignment A confident assignment is indicated in the file with a symbol see Output File section One may wish to first edit this file to remove some symbols see Advice section The tolerance for chemical shift matching follows the rules discussed above Enter the desired matching tolerance and weight factor in the appropriate field Matching old assignment Chemical shift tolerances sigma in ppm 0 30 0 30 foan 0 30 0 30 Weighting factors must be integers for old assignments COold CAald CBold ae ao a 14 8 Simulated Annealing Parameters The lower section of the GUI is used to set the simulated annealing schedule The annealing schedule may have a single or multiple steps To add or delete a step click Add a row or Delete last row respectively Simulated annealing parameters istart istep tin swap_lim gamma swap_sz noe_scale 2 ca Ha 1 200 poo foo o fiso o 20000 ja E B p p P 2 EN foo 10 0 so000 5 aa i moor pa fo fio fa Add a row Delete last row 8 1 Annealing Schedule Each Annealing schedule step has several user adjustable features 8 11 Number of cycles This specifies the number of times or cycles the annealing schedule will be run Since the initial assignments are random these will be independent solutions 8 12 tstart tstep tfin The parameter tstart defines the starting temperature for that annealing step A
26. nt are 3D cross peaks between the current problem and a related solved problem 1 e unliganded and ligand bound forms of the same protein Following the initial scoring two spin systems or two blocks of spin systems are chosen at random and swapped The new solution 1s scored based upon the same criteria If the swap results in an improvement in the score the swap is accepted If the swap results in a poorer score the difference in score is compared to the temperature of the system and the swap may or may not be accepted At infinite high temperature all swaps are accepted As the temperature of the system is lowered simulated annealing swaps that result in an improvement in the overall score will be preferentially accepted Monte is usually executed for 10 to 50 cycles starting from different random starting positions The best solution from each of these independent cycles are compared in order to evaluate the reliability of an assignment 2 License Agreement Academic and non profit institutions are welcome to use Monte at no cost However users are required to sign an license agreement that is available on the program web site Return the two signed copies of the license agreement to Dr Gordon S Rule Department of Biological Sciences 4400 Fifth Avenue Carnegie Mellon University Pittsburgh PA 15213 After receipt of the license agreement you will be emailed a license key This license key must by place
27. ntly Monte only takes advantage of amide amide NOE data Hence the HA and HB fields are blank 17 9 3 Text Output At the end of each annealing step as well as the initial random assignment Monte writes a text file to the experimental subdirectory of the current best assignment solution The text files have the name filename_out 00 filename_out 0l etc where filename is the root filename of the experimental data The best solution from all of the different runs 1s placed in the fine filename_final The format of the text output file is indicated by the following example 202 PRO ON 202 AU CL ZOE ASN AR 2D LOL 209 GEX oe 0 O00 0 0 0 00 O 0 0 ZUGr AON OZ 0200 Lo 07 0 0 lt 0 Zee OO 00 0 00 Ola 00 0 00 0 00 0 00 0 00 A MALO DTO CIAO LO O LIADO ae OO sok OOe2 OSs82 29x Se Saad LT Seeds nS ae Z 00 000 0200 0 00 0 00 0 00 0 00 42 48 00 43 08 35 83 1 00 LEIOA LAA SLO E266 Ad LADO DO FIAS OUA is oF Dl A Hasdo LO Spry Lor LISA IIS AO LI li ADS LIGA E 4 LIE GLY LOV ZOT L06779 ZOO Lro VOD 208 LL39 209r GEN LLO O lo E ca hl DO Own 7IO WO OW O o Ae o Ae ao ao Peaks not used ao ao pa OY LEO Qs Adios kk 03 9290 Jol OL gol 0 00 OSO 0 00 ds 00 The first field in each row 1s the residue number This number may be followed by a The plus symbol indicated that the reported spin system was assigned that residue number in all of the cycles For example if 10 cycles w
28. oach is provided in the rho_dna directory that is included in the distribution This feature is useful when working on chemical shift assignments of a particular protein and subsequent ligand bound forms of the same protein The assumption here is that the presence of the ligand only significantly perturbs chemical shifts of residues that are in the vicinity of the binding site or are involved in structural changed due to the ligand binding In this case many of the chemical shifts obtained for the unliganded form of the protein provide a useful constraint for assignment of the ligand bound form of the protein or vice versa To match an old assignment solution Monte compares how well cross peaks in the same type of 3D spectra match Monte will match all or some of HNCO HN CA CO HNCA HN CO CA HNCB and HN CO CB types data To match an old output select match old output One must also tell Monte what columns to find the desired E match old output Open old CS match table data These column definitions identified in a manner IN similar to that described above See Chemical Shift Definitions Open the old CS match table to set the column definitions The user must also provide the path and filename of the old Monte output file One may bowse the directory structure for the desired file Old table filename include path rho rho_old out Browse Monte will only match chemical shift assignments in this old output file that are i
29. of the Gaussian function that was used in scoring hence all have the same thickness The rows that display the information from NOE data show consecutive NOEs i 1 1 local NOEs as well as long range NOEs i 1 gt 4 An open circle indicates that the NOE was predicted from the input structure either primary or tertiary A filled circle indicates that the NOE was found in the experimental data Where possible filled circles are connected by a solid line between the two coupled amide protons For example panel A shows long range NOEs between residues 3 and 4 and 55 and 56 respectively In the three dimensional structure these residues are across from each other on a P sheet If it is not possible to connect the coupled residues by a line because they reside on separate lines in the output figure then the NOE partner is indicated by the residue number For example Argll shows an NOE to Asn204 The last line of this figure provides information on the uncertainty of the assignments A bar is printed under each residue with a height that is proportional to the number of different spin systems that were assigned to this particular residue in the independent trials A zero height bar thin line indicates that the same solution was found in all of the independent trials For example Tyr3 Thr4 Vall0 Argl1 Gly12 Cys14 and Alal5 appear to be uniquely assigned with this data In contrast residues Val5 Phe8 are not assigned for reasons discussed in the tex
30. olution If a best fit solution is not found for a spin system it will listed at the end of this text file under peaks not used This line is preceded by so that if one would like to use this output for matching chemical shifts these lines will be ignored when read by Monte 9 4 HTML To help in visualizing the assignment solution and determining the quality of the assignment solution at the end of a run Monte writes four html files Use an Internet browser to open the main html file in the experimental subdirectory This html document will set up several frames on you browser 18 The top frame contains the assignment information At the top of this frame there 1s a color code key which indicates the colors that highlight chemical shift mismatches shades of yellow indicating mismatches that are greater than times the set tolerance greater than 2 times the tolerance and greater than 3 times the set tolerance The next row lists the column headers The residue number and residue type are taken from the pdb file or sequence file The next columns are the peak number and assignment number if found in the cs file These columns are followed by the chemical shifts atom types that are identified in the par file gt 1x gt 2x 53x tolerance ZZ res type peak asn H N Ca Caf CaCO CO Amide NOEs Shown below is an example of the body of this html frame In this example the CA matching tolerance was set to
31. own on the right Fail al 1433 Kh h at h d h 4 d h bhtttaid bak k 143 FRA a CO Jaa A gt SS om r os Mes Iana Uncertainty 0O MO IA _ ___ g____O______ CA HOE 70 t 100 17 12 J TLYQSNTILAHLGATLOLYGEDOOG EA GWEDLACEYTISL FT Y EAG ED Du KE CE LJ ta IA LEAL a ut LAA LA AA iat ae i NOE 3 4 gt 4 Uncertainty OO __ lieO___ __ N 20 The amino acid sequence is given at the top of each row The half circle symbol under two adjacent residues indicates that this assignment agrees with information from specific C 1 labeling For example the amide of Thr4 was identified as being coupled to a C carbonyl in a sample that was uniformly labeled with N and specifically labeled with C 1 Tyr An underlined residue indicates that the assignment is consistent with specific N labeling For example the amide resonance for the spin system that was assigned to Val10 was present in a sample labeled with N Val The following line indicates the secondary structure either provided by the user or calculated from the three dimensional structure of the protein The next three lines labeled with J indicate the fidelity of matching of inter and intra residue chemical shifts The thickness of the bar is related to how well the chemical shifts match Thinner lines indicate poorer matches such as the Jcg coupling between residues 37 and 38 panel A In panel B all of the indicated matches are within one o
32. r this information alone is often too general to provide a useful constraint for finding a unique chemical shift assignment solution Therefore it is highly recommended that the complementary pair of information is used for input The format of the 3D NOESY files 1s similar to the 4D input file described above however in these cases the parent amide shifts are followed by the single proton or nitrogen cross peak For example 3d H NH HSQC NOESY PARENT NOE 1 NOE 2 NOE 3 1000 124 2 7205 8 54 1 22 9 00 13 N NH HSOC NOESY PARENT NOE 1 NOE 2 NOE 3 MO OO TA Techo 23 23 120 4 128 4 5 Parameter File 5 1 Overview The parameter par file is unique to each chemical shift assignment project and must reside in the assignment project sub directory This is a plain text file however it is strongly suggested that this file only be edited via the graphical user interface fm tcl The parameter file provides Monte with several pieces of information such as the root filename for the parameter data and output files In addition the directory path for the assignment problem is also in the parameter file The parameter file also identifies to Monte what data files are available for input the column definitions of the chemical shift data base cs file and the parameters for chemical shift matching as well as weighting factors for different scoring criteria When one launches fm tcl three windows will appear the main
33. r interation as the temperature is lowered The parameter gamma defines the rate of increase of ncyclim as temperature is lowered for a particular annealing step The desired result is to normalize the number of successful swaps over the course of the annealing step 15 8 14 swap_sz The parameter n_swap defines the maximum number of adjacent spin systems that will be swapped during that annealing step The size of the blocks that are swapped is chosen at random from 1 to n_swap Initially n_swap should be low 1 e 1 since it is unlikely that the random assignment resulted in correctly assembling linked spin systems As the temperature is lowered more spin systems will be better linked and the problem becomes finding the correct position in the primary sequence of the protein Therefore at later stages of the annealing schedule it may be advantageous to swap larger blocks of spin systems 8 15 noe_scale Matching predicted NOE s from a pdb file to experimental NOE s is a very powerful assignment tool This information becomes most useful to Monte after some spin systems become linked and better assignment In addition since Monte internally scores all perfect matches to 100 and a single parent amide may have multiple NOE s NOE matching may over power J coupled matching For these reasons the NOE scaling factor may be adjusted at each annealing step Also in contrast to other scaling factors the NOE scaling factor may be less than
34. r the assignment project Root Filename sample When the filename in the Root Filename field above is changed all the expected filenames for the next section are updated This section of the GUI also allows the user to select the level of information that is written to the filename log file The highest log level 3 should be used for diagnostic purposes A log level of one is satisfactory in most situations In order for the HTML Corr Plot and Solution Output buttons to work properly the names of the appropriate applications have to be included in the fm tcl file See the last section of this guide Next identify what files are available for the Monte run by clicking on the appropriate radio button see below Files avallable E sample cs ve Sample seg sequence file sample pdb pdb file E sample ss secondary structure file JJ sample 3dh_ noe 3D H HH NOESY _ sample ddnoe 4D HA NH NOESY _ sample csl J sample nsi The cs file MUST be available Either the seq or pdb file MUST also be available The other files are optional One may only use the N NH HSQC NOESY data if there is also H HN HSQC NOESY data available hence in the above example the 3dn_noe choice may not be selected 5 3 Output Files Monte produces the following output files main html An HTML document summarizing chemical shift matching NOE matching and residue type prediction filename_out The best ass
35. shift that is 3 times the matching tolerance If the Gaussian extends past 1 00 ppm All shifts differences that are 1 00 ppm or greater are set to a score equal to the baseline value In some cases the baseline may be set below zero see Annealing Schedule section on Repulsive Terms An example of the distribution for scoring is shown in Table 1 11 5 45 Chemical Shift offsets To determine reside type for the i and 1 1 residues Monte compares the input chemical Chemical shift offsets for amino acid type determination shifts to the median shifts for each residue type ca off ch off co_off cy_off and secondary structure type found in the aaconv 10 0 Fi ss 0 0 file see aaconv section One s chemical shift referencing may differ from the standard shifts In this case one may instruct Monte to apply a correction to the input data for residue type identification Enter the values in the appropriate boxes provided These corrections are only applied when determining residue type probabilities and Monte will not actually change the chemical shifts between the input and output files Table 1 Example of distribution of scoring for chemical shift matching for three different Guassian tolerence widths The chemical shift difference between complementary shifts 1s indicated in the first column ppm In this example a tolerence is 0 3 ppm for CO 0 15 ppm for CA and 0 05 ppm for HA ppm CO Ca Ha ppm CO Ca Ha ppm CO Ca Ha PSB LOG LO 0
36. t Open rectangles mark the location of Pro residues 20 9 6 Correlation Plot Additional detail on the existence of alternative Residue assignments 1s presented in a correlation plot This is a 20 40 60 so 100 120 140 160 180 200 postscript file that is found in the solutions directory Figure Legend The x axis indicates the residue number and the y axis indicates to what other residues besides that found in the best solution the spin system was assigned to For example the circled point in the plot indicates that the spin system that was most frequently assigned to residue 88 is also assigned to 179 in one or more of the 20 solutions The intensity of the plotted point is proportional to the frequency that a spin system is assigned to a particular residue The more darkly colored the point the higher the frequency of the assignment of a single spin system to a particular residue For example the dark diagonal in this plot shows that most spin systems are uniquely assigned to a single residue The rectangular area below the lower horizontal line represents the cache area Points found in this region indicate that spin system was placed in the cache 1 e unassigned in some or all of the assignment solutions Alternate Assignment 10 Customizing and Running Monte 10 1 Customizing If you desire to use the buttons to launch the browser and postscript viewers for viewing the output files then it is necessary to tell Monte
37. ta found in these files is used by Monte to determine amino acid type probabilities for the 1 and i 1 residues of a amide pair spin system One may wish to view this file to determine if a chemical shift offset should be added to or subtracted from the user input shifts N CO CA CB or CG shifts for the purpose residue type calculations See Parameter file section A slight difference in chemical shift references may affect the assignment of residue type 4 3 Chemical shift Data Base The Chemical shift data base contains all J correlated chemical shifts for a particular spin system A spin system 1s defined here as all chemical shifts that may be correlated to a particular amide nitrogen proton pair These chemical shifts are used to both identify the residue type of the i and i 1 residues as well as linking spin systems by matching complementary pairs of chemical shifts Chemical shifts may include data collected from any or all of these types of experiments HNCA HN CO CA HNHA HA CACO NH HNCACB CBCACONH HNHB HBHA CBCACONH HNCO HN CA CO H N COCA NH HN CACB CG HN COCACB CG H N CACO NH The format of the chemical shifts data base data follows the following example N HN CA 1 CA a L HATA HAZ 2 HALA SL HAZ a 1 es LOOO GEULZ 2a 2S TIOS DO 60454 3 43 Ola 00 O 00 OOU isos LUTO GLT TO bast Des 0400 43 41 2 43 000 Sey ae Care ea ere rer Any line that is started by either or is considered a comment te
38. window and one chemical shift definition window An additional chemical shift definition window for utilizing existing chemical shifts will be opened if this option is selected If the tcl code does not run make sure the lastest versions of tcl tk are installed and the symbolic links are set see Getting Started The main window is divided into two parts The upper part contains fields for setting general parameters and the lower part contains fields for setting the simulated annealing parameters The very bottom of the man window contains a series of buttons to allow you to control a number of processes HTML Corr Plot Solution Output Open ES ID table Open par file Sawe par file Stark Kill Quit The function of these buttons is as follows e HTML Launches a browser to view the HTML output file e Corr Plot Launches a postscript viewer to view the correlation plot e Solution Output Launches a postscript viewer to view the summary output e Open CS ID table Opens the table that identifies the atom types in the input chemical shift file e Open par file Opens a parameter file e Save par file Saves the parameter file e Start Runs Monte prompts for a parameter file e Kill Terminates current run of Monte e Quit Closes the tcl window 5 2 Input Files At the top of the GUI there are two fields in which one should enter the root filename for the assignment problem and the subdirectory path Pathname fo
39. xt and will be ignored The first column in any data entry is a peak identification number This is a four digit number with one decimal place The last digit may be incremented for example to indicated a single parent amide resonance that may give rise to more than one carbon or proton cross peaks 1 e from minor protein conformations or overlapping resonance signals The second entry is an optional comment of up to eight characters If there is a number in this field the software will assume that this is a tentative assignment This number is not used in the Monte Carlo protocol however is carried over to the output files Using this field to add change tentative assignment is useful in comparing output from multiple Monte runs Subsequent entries in each row are chemical shift frequencies and may be separated by spaces or comas The first and second columns must contain the parent amide nitrogen and amide proton shifts in either order defined by the par file The following chemical shifts may be in any order but must be consistent throughout the table The order is user defined in the par file see section on Parameter File Any chemical shifts at the very end of the list may be left blank if the shift is unknown However a unknown chemical shift field positioned within a series of known shifts must be entered 0 00 to hold its place see example peak 101 0 above no shift is identified with the intra residue CA 4 4 Sequence File
Download Pdf Manuals
Related Search