Automated calculation of a protein-ligand complex structure (AUREMN, Brazil 2018): Difference between revisions

From CYANA Wiki
Jump to navigation Jump to search
No edit summary
No edit summary
Line 79: Line 79:
Resonance assignment within cyana is done using the module FLYA. Spectra are processed and referenced relative to each other. Then, peak lists are generated and cleaned as much as possible (water peaks, noise peaks etc.). In order to read the prepared peak lists, the spectrum definitions have to be added to the xeasy peak list as in exercise 1.
Resonance assignment within cyana is done using the module FLYA. Spectra are processed and referenced relative to each other. Then, peak lists are generated and cleaned as much as possible (water peaks, noise peaks etc.). In order to read the prepared peak lists, the spectrum definitions have to be added to the xeasy peak list as in exercise 1.


== Experimental input data ==
=== Experimental input data ===


In the most general sense, there are two type of experiments used for protein resonance assignments.
In the most general sense, there are two type of experiments used for protein resonance assignments.

Revision as of 14:56, 25 February 2018

In this tutorial we will determine the resonance assignments and the structure of a protein-ligand complex using modules of CYANA.

To this end we will first run the CYANA module FLYA to obtain the resonance assignments from backbone, side chain and NOESY experiments (actually, the xeasy peak lists of these experiments).

Then we will use noeassign to assign the NOESY spectra and calculate the holo protein structure without the ligand.

In a next step we will first draw the ligand, convert the obtained SMILES code to a *.mol2 file and generate the *.lib file for CYANA.

Then we will assign intermolecular peaks lists and redo the structure calculation, this time of the protein-ligand complex.

To finalize you will compare the calculated NMR structure to an Xray structure and generate statistics.

And ultimately you can try to improve your structure results by studying and applying the options available within the FLYA and noeassign modules of CYANA.


CYANA setup for the AUREMN Practical NMR Course in Campino (24-26 February 2018)

Please follow the following steps carefully (exact Linux commands are given below; you may copy them to a terminal):

  1. Go to your home directory (or data directory).
  2. Get the data for the practical from the server (AUREMN2018.tgz).
  3. Unpack the input data for the practical.
  4. Get the demo version of CYANA for this practical.
  5. Unpack CYANA.
  6. Setup the CYANA environment variables.
  7. Change into the newly created directory 'AUREMN2018'.
  8. Copy the demo_data directory to 'flyabb'.
  9. Change into the subdirectory 'flyabb'.
  10. Test whether CYANA can be started by typing its name, 'cyana'.
  11. Exit from CYANA by typing 'q' or 'quit'.
  12. Download Chimera (to your personal laptop) from: Chimera
  13. Download Avogadro (to your personal laptop) from: Avogadro
cd ~
cp /home/julien/AUREMN2018.tar.gz .
tar zxf AUREMN2018.tar.gz 

wget 'http://www.cyana.org/wiki/images/6/64/Cyana-3.98bin-180213Demo.tgz'
tar zxf Cyana-3.98bin-180213Demo.tgz
cd cyana-3.98/
./setup

cd ~

cd AUREMN2018
cp -r demo_data flyabb 
cd flyabb


cyana
___________________________________________________________________

CYANA 3.98 (mac-intel)

Copyright (c) 2002-17 Peter Guentert. All rights reserved.
___________________________________________________________________

    Demo license valid for specific sequences until 2018-12-31

    Library file "/Users/deans/cyana-3.98/lib/cyana.lib" read, 41 residue types.
*** ERROR: Illegal residue name "LIG".
*** ERROR: Cannot read line 114:
           LIG  333
cyana> q

If all worked, you are ready to go in terms of everything related to CYANA! The reason you see the ERROR message, is because you have a sequence file in the directory, but not library file for the ligand yet. Don't worry, this is as expected and you will take care of it during the exercise.

If you want to return to your practical later, using your own Linux or Mac OS X computer, you can download the demo version of CYANA from here.

Hint: More information on the CYANA commands etc. is in the CYANA 3.0 Reference Manual.

Automated resonance assignment

Resonance assignment within cyana is done using the module FLYA. Spectra are processed and referenced relative to each other. Then, peak lists are generated and cleaned as much as possible (water peaks, noise peaks etc.). In order to read the prepared peak lists, the spectrum definitions have to be added to the xeasy peak list as in exercise 1.

Experimental input data

In the most general sense, there are two type of experiments used for protein resonance assignments. Through bond, TOSCY kind of experiments and through space NOESY type of experiments. Each of these two experiments carries distinct information that help the resonance assignment. The HSQC, HMQC or TROSY elements of these experiments merely help the resolution, by allowing the separation of resonances according to spin types (1H, 13C, 15N) into additional dimensions.

At the very minimum, for small systems and in favorable cases, a NOESY experiment may be sufficient to get an assignment and enough distance restraints for a structure calculation.

Peak lists in XEASY format are prepared by automatic peak picking with a visualization program such as CcpNmr Analysis, NMRdraw or NMRview are stored in files XXX.peaks, where XXX denotes the FLYA spectrum type. Then they are cleaned (unnecessary water and noise peaks removed).

As part of the supplied data for the exercises, experimental peak lists are available for the following spectra:

  • HNtrosy (spectrum type 'N15HSQC' in the CYANA library)
  • trHNCA (spectrum type 'HNCA' in the CYANA library)
  • HNCOCA (spectrum type 'HNcoCA' in the CYANA library)
  • HNCACB (spectrum type 'CBCANH' in the CYANA library)
  • HCCCHTOCSY (the spectrum type will have to be determined in the first exercise)
  • NTOCSY (spectrum type 'N15TOCSY' in the CYANA library)
  • 3D [13C]-resolved NOESY called aro (spectrum type 'C13NOESY' in the CYANA library)
  • 3D [13C]-resolved NOESY called cnoesy (spectrum type 'C13NOESY' in the CYANA library)
  • 3D [15N]-resolved NOESY called nnoesy (spectrum type 'N15NOESY' in the CYANA library)

Each peak list starts with a header that defines the experiment type and the order of dimensions. For instance, for HNCA.peaks:

# Number of dimensions 3
#FORMAT xeasy3D
#INAME 1 HN
#INAME 2 C
#INAME 3 N
#SPECTRUM HNCA HN C N
      5   6.475  58.033  98.548 1 U   2.769E+02  0.000E+00 e 0     0     0     0
      6   6.476  62.123  98.126 1 U   2.571E+01  0.000E+00 e 0     0     0     0
      7   6.475  54.017  98.159 1 U   2.547E+01  0.000E+00 e 0     0     0     0

The first line specifies the number of dimensions (3 in this case). The '#SPECTRUM' (no space between characters) lines gives the experiment type (HNCA, which refers to the corresponding experiment definition in the CYANA library), followed by an identifier for each dimension of the peak list (HN C N) that specifies which chemical shift is stored in the corresponding dimension of the peak list. The experiment type and identifiers must correspond to an experiment definition in the general CYANA library (see below). If a definition is missing for an experiment it must be added to the CYANA library. After the '#SPECTRUM' line follows one line for every peak. For example, the first peak in the 'HNCA.peaks' list has

  • Peak number 5
  • HN chemical shift 6.475 ppm
  • C (CA) chemical shift 58.033 ppm
  • N chemical shift 98.548 ppm

The other data are irrelevant for automated chemical shift assignment with FLYA. In particular, the peak volume or intensity (2.769E+02) is not used by the algorithm.

Hint: The formats of other CYANA files are described in the CYANA 3.0 Reference Manual.


The protein sequence is supplied by three-letter code in a XXX.seq file.

As part of the supplied data for the exercises there are two sequences:

  • demoShort.seq (the protein sequence alone)
  • demoLong.seq (the protein sequence, ligand and a linker that connects the two molecules)

Linker sequences serve to keep two or more molecules close in coordinate space during calculations, is usually between 15-20 elements long and is composed of dummy atoms that allow the linking.

SPECTRUM definitions in the CYANA library

When you start CYANA, the program reads the library and displays the full path name of the library file. You can open the standard library file to inspect, for example, the NMR experiment definitions that define which expected peaks are generated by FLYA. For instance, the definition for the HNCA spectrum (search for 'HNCA' in the library file 'cyana.lib') is

SPECTRUM HNCA  HN N C
 0.980  HN:H_AMI  N:N_AM*  C:C_ALI  C_BYL
 0.800  HN:H_AMI  N:N_AMI  (C_ALI) C_BYL  C:C_ALI 

The first line corresponds to the '#SPECTRUM' line in the peak list. It specifies the experiment name and identifies the atoms that are detected in each dimension of the spectrum. The number of identifiers defines the dimensionality of the experiment (3 in case of HNCA).

Each line below defines a (formal) magnetization transfer pathway that gives rise to an expected peak. in the case of HNCA there are two lines, corresponding to the intraresidual and sequential peak. For instance, the definition for the intraresidual peak starts with the probability to observe the peak (0.980), followed by a series of atom types, e.g. H_AMI for amide proton etc. An expected peak is generated for each molecular fragment in which these atom types occur connected by single covalent bonds. The atoms whose chemical shifts appear in the spectrum are identified by their labels followed by ':', e.g. for HNCA 'HN:', 'N:', and 'C:'.

Exercise 1: Determine the spectrum type

For the HCCCHToscy, determine the spectrum type and put the definition in the HCCCHTocsy.peaks file with the appropriate syntax.

The experiment is a TOCSY, a through-bond experiment. It allows you to see, in this case, from the backbone all the way out into the side chains.

  • Use the less command (to view files in the terminal but not change) to search the spectrum type in the 'cyana.lib' file.

Hint: Look at the definitions themselves and not just the SPECTRUM names, to determine which TOCSY is the appropriate one. Take the experiment with the most through-bond transfers.

  • work in the copy of the data directory ('cd flyabb')
  • Use a graphic text editor (or if you feel comfortable, use the vi terminal text editor) to manipulate the HCCCHTOCSY.peaks file and enter the appropriate spectrum type and identifiers in the correct order.
  • Check that the order of the dimensions in your SPECTRUM definition matches the actual experiment.

Depending on how the xeasy peak file was generated, the order of the dimensions does not have to match the way experiments are recorded,

and of course they do not necessarily match the SPECTRUM definition given in the 'cyana.lib' file.

Hint: A quick determination of the order of the dimensions and atom types, can be done by looking at the columns of the chemical shifts and detect the chemical shift patterns.

This may be harder than it sounds at first instance, take your time to detect the pattern if it is not immediately obvious to you.

As you know, the chemical shifts of specific atom types and groups are quite distinct. If you need help with the chemical shift statistics, go to: BMRB


Hint: For information on how to use the vi terminal editor: vi editor

Exercise 2: Run FLYA

  • work in the copy of the data directory ('cd flyabb')

Using the text editor of your choice, create your 'init.cya' macro as outlined (The init macro) and also your 'CALC.cya' macro (The FLYA CALC macro) to run FLYA. Be extra careful to avoid typos and unwanted spaces in coma lists etc.

Execution scripts or "macros" in CYANA

For more complex task within CYANA, rather than to enter the execution commands line by line at the CYANA prompt, the necessary commands are collected in a file named '*.cya'.

Therefore, CYANA scripts ("macros") 'CALC*.cya' contain the various commands to perform the required tasks.

The init macro

One has the fixed name 'init.cya' and is executed automatically each time CYANA is started. It can also be called any time one wants to reinitialize the program. It contains normally at least two commands that read the CYANA library and the protein sequence:

rmsdrange:=15-111
cyanalib
read demoShort.seq

The first line sets the appropriate rmsdrange, the command 'cyanalib' reads the standard CYANA library. The next command reads the protein sequence.

The protein sequence is stored in three-letter code in the file 'demo.seq'.

The FLYA CALC macro

The 'CALC.cya' starts with the specification of the names of the input peak lists:

peaks:=aro,cnoesy,nnoesy,HNtrosy,trHNCA,HNCOCA,HNCACB,NTocsy,HCCCHTocsy

The peak list names are separated by commas (without blanks!). The files on disk have the file name extension .peaks, e.g. 'HNCA.peaks'.

The commands above will use all available peak lists. You can choose any subset of them by modifying the 'peaks:=...' statement.

These are followed by tolerances for chemical shift matching:

assigncs_accH=0.03
assigncs_accC=0.4
assigncs_accN=assigncs_accC
tolerance:=$assigncs_accH,$assigncs_accH,$assigncs_accC

In this case, a tolerance of 0.03 ppm will be used for protons, and 0.4 ppm for carbon and nitrogen. The 'assigncs_accX' variables are used within flya, the tolerance variable is used in consolidation.

The next parameter specifies the seed value for the random number generator (an arbitrary positive integer is ok).

randomseed=101

The next parameter is set in order to speed up the calculation for this practical:

shiftassign_quick=.true.

In production runs, better results can be expected (at the expense of longer computation times) if this parameter is not set. The parameters specifies:

  • An option to choose the "quick" optimization schedule.
shifts_consolidate_swap=.true.

Swapping is optimized in comparison to the supplied reference assignment (only applied during consolidation).

Finally, there is the command to start the FLYA algorithm:

flya runs=10 assignpeaks=$peaks shiftreference=manREF.prot

Here, the given parameters of the 'flya' command specify that:

  • The number of independent runs of the algorithm, from which the consolidated shift will be calculated (chosen smaller than in normal production runs in order to speed up the calculation).
  • The input peak lists that will be used (as defined above).
  • An ensemble of random structures will be calculated for generating expected peaks (leads to prediction of short range NOES in NOESY-type experiments).
  • The results will be compared with the reference chemical shifts in the file 'manREF.prot' (which have been determined by conventional methods).


When you have prepared the 'init.cya' and the 'CALC.cya' start your FLYA ('CALC.cya') macro using 10 processors by calling it as outlined just below. It will take between 10-20 minutes (depending on your system) to complete the assignment, once the calculation starts.

To run the FLYA calculation, one could start CYANA and execute the 'CALC.cya' macro from the CYANA prompt, however on a computer with multiple processors it is better to speed up the calculation by running the the 'CALC.cya' macro through a MPI scheduler:

cyana -n 10 CALC.cya

This starts 10 independent calculations on 10 processors by using the MPI scheduler (if installed on your system, otherwise shared memory will be used).

It is strongly recommended that you check the MPI scheduler if your calculation is running. If you made a mistake in one of the two macros, the calculation may or may not start, or get interrupted at some point.

ls -ltr 

This command allows you to check if the calculation has generated intermediary files or final results, with a time stamp.


To check the general load on your local computer use:

top

To kill all processes running (from you):

skill -u <username>

FLYA output files

The FLYA algorithm will produce the following output files:

  • flya.prot: Consensus assigned chemical shifts. This file contains a chemical shift for every atom that has been assigned to least one peak.
  • flya.tab: Table with details about the chemical shift assignment of each atom (comparison with reference shifts). In this file you can see for each atom whether the assignment is "strong" (self-consistent) or "weak" (only tentative).
  • flya.txt: Assignment statistics
  • flya.pdf: Graphical representation of the assignment results
  • XXX_exp.peaks: List of expected peaks, corresponding to input peak list XXX.peaks
  • XXX_asn.peaks: Assigned peak list, corresponding to input peak list XXX.peaks

The flya.txt file

This output file starts with overall assignment statistics for each group of atoms as defined by 'analyzeassign_group:=...':

   ____________________________________________________________

   CHEMICAL SHIFT ASSIGNMENT
   ____________________________________________________________

   SEED: 1
   chemical shifts for 542  atoms found
   Peaks assigned from frequencies

   BB: REFERENCES(2):512 CHEMICALSHIFTS(1):542 (1)and(2):512 MATCH:507(99.0% of (2))
  • REFERENCES(2) is the number of reference assignments (in the selected group)
  • CHEMICALSHIFTS(1) is is the number of atoms assigned by FLYA
  • (1)and(2) is the number of atoms that are assigned by FLYA and in the reference.
  • MATCH is the number of atoms with the same assignment by FLYA and in the reference. The percentage is relative to the number of reference assignments.

Further below comes a table with information about each peak list:

   PEAKLISTS
   #Expected: Total number of expected peaks
   noRef: Number of expected peaks with missing reference shifts
   noPeak: Number of expected peaks for wich no peak can be measured
   Assigned: Number of expected peaks that could be assigned
   Match: Number of assigned peaks that fit reference shifts
   #Measured: Total number of peaks in peak list
   Assigned: Number of measured peaks that could be assigned to expected peaks
   exp/meas: Ratio of assigned expected and measured peaks

   Lists      #Expected  noRef   noPeak   Assigned        Match    #Measured Assigned  exp/meas Assigned
   N15HSQC        106       8       1   104( 98.11%)    97( 91.51%)    131     96( 73.28%)   1.1
   HNCA           211      15      11   194( 91.94%)   186( 88.15%)    329    179( 54.41%)   1.1
   HNcaCO         211      15      11   197( 93.36%)   183( 86.73%)    246    176( 71.54%)   1.1
   HNCO           105       7       1   101( 96.19%)    97( 92.38%)    158     97( 61.39%)   1.0
   HNcoCA         105       7       0   101( 96.19%)    97( 92.38%)    158     99( 62.66%)   1.0
   CBCANH         399      26      25   361( 90.48%)   350( 87.72%)    623    339( 54.41%)   1.1
   CBCAcoNH       200      13       2   196( 98.00%)   185( 92.50%)    324    192( 59.26%)   1.0
   ALL           1337      91      51  1254( 93.79%)  1195( 89.38%)   1969   1178( 59.83%)   1.1

It contains the following data:

  • #Expected: Total number of expected peaks
  • noRef: Number of expected peaks with missing reference shifts
  • noPeak: Number of expected peaks for which no peak can be measured
  • Assigned: Number of expected peaks that could be assigned based on the reference chemical shift assignments. The theoretical maximum of 100% corresponds to the situation that the spectra “explain” all expected peaks. Each expected peak can be mapped to at most one measured peak. Remaining expected peaks correspond to missing peaks in the measured peak list.
  • Match: Number of assigned peaks that fit (within tolerance) reference shifts. The theoretical maximum of 100% corresponds to having all measured peaks assigned. Note that several expected peaks can be mapped to the same measured peak, i.e. the assignments of measured peaks can be unambiguous or ambiguous. Remaining unassigned measured peaks are likely to be artifacts.
  • #Measured: Total number of peaks in peak list
  • Assigned: Number of measured peaks that could be assigned to expected peaks
  • exp/meas: Ratio of assigned expected and measured peaks

There is more information on the results of the assignment calculation in the 'flya.txt' file (not described here).

The flya.tab file

This file provides information about the chemical shift assignment of each individual atom:

   Atom  Residue      Ref   Shift     Dev  Extent  inside   inref
   ...
   N     GLY   57 102.109 102.043   0.066    10.0   100.0   100.0  strong=
   H     GLY   57   8.571   8.570   0.001    10.0   100.0   100.0  strong=
   CA    GLY   57  45.415  45.433  -0.018    10.0   100.0   100.0  strong=
   HA2   GLY   57   4.042
   HA3   GLY   57   3.436
   C     GLY   57 173.621 173.662  -0.041    10.0    89.4    90.0  strong=
   N     LEU   58 120.640 120.649  -0.009    10.0    80.0    80.0  =
   H     LEU   58   7.488   7.492  -0.004    10.0    79.8    80.0  =
   CA    LEU   58  51.943  51.940   0.003    10.0    70.0    70.0  =
   HA    LEU   58   4.995
   CB    LEU   58  45.602  45.568   0.034    10.0    82.7    80.0  strong=
   CG    LEU   58  26.528
   HG    LEU   58   1.515
   CD1   LEU   58  24.745
   C     LEU   58 173.619 174.576  -0.957    10.0    40.1    10.0  ! (C 59)
   ...
  • Ref: Chemical shift value in the reference chemical shift list (ref.prot). It was not used in the calculation.
  • Shift: Consensus chemical shift value from FLYA
  • Dev = Ref - Shift
  • Extent: Number of runs in which the atom was assigned by FLYA.
  • Inside: Percentage of chemical shift values from the (10) independent runs of FLYA that agree (within the tolerance) with the consensus value.
  • inref: Percentage of chemical shift values from the (10) independent runs of FLYA that agree (within the tolerance) with the reference value.
  • Outcome of the assignment:
    • strong: "strong" assignment, i.e. Inside > 80%.
    • =: Assignment that agrees with reference, i.e. Dev < tolerance.
    • !: Assignment that does not agree with the reference, i.e. Dev > tolerance.
    • (atom name): Correct assignment, if within the same residue (no residue number given), or the neighboring residues.

The flya.pdf file

This PDF file provides a graphical representation of the 'flya.tab' file. Each assignment for an atom is represented by a colored rectangle.

flya.pdf
  • Green: Assignment by FLYA agrees with the manually determined reference assignment (within tolerance)
  • Red: Assignment by FLYA does not agree with the manually determined reference assignment
  • Blue: Assigned by FLYA but no reference available
  • Black: With reference assignment but not assigned by FLYA.

Respective light colors indicate assignments not classified as strong by the chemical shift consolidation. The row labeled HN/Hα shows for each residue HN on the left and Hα in the center. The N/Cα/C’ row shows for each residue the N, Cα, and C’ assignments from left to right. The rows β-η show the side-chain assignments for the heavy atoms in the center and hydrogen atoms to the left and right. In the case of branched side-chains, the corresponding row is split into an upper part for one branch and a lower part for the other branch.

Exercise 3: Analyze the FLYA results

  • Analyze your FLYA results using 'less' or a graphical editor and a pdf viewer.
  • What do you think?
  • How does the automated assignment compare to the provided assignment?
  • How robust you think are the results?
  • What could you do to likely improve the result?

Hint: Use the terminal command 'gs' to view pdf files (control-C to quit gracefully):

gs flya.pdf

Using Talos to generate aco restraints

Angle restraints from the backbone chemical shifts help restrict angular conformational space. We wish to use only "strong assignments" to generate these restraints.

If you do not have TALOS installed get it from here. It is part of the nmrpipe software package.

Exercise 4: Calculate the backbone angels using Talos

Hint: Copy the FLYA results into a new folder, since otherwise you will overwrite your original 'flya.prot' file.

Essentially you will need to copy the details directory and the 'flya.prot' file.

cp -r  flyabb acoPREP
cd acoPREP
rm *.peaks *.out *.job

Use the text editor of your choice, to create a 'CALC.cya' file with the commands to calculate the talos angles.

TALOS is used to generate torsion angle restraints from the backbone chemical shifts in 'flya.prot'.

consolidate reference=flya.prot file=flya.tab plot=flya.pdf prot=details/a[0-9][0-9][0-9].prot

This overwrites the original flya.pdf with only strong assignments.


read prot flya-strong.prot unknown=skip

talos talos=talos+                
talosaco pred.tab

write aco talos.aco

This will call the program TALOS and store the resulting torsion angle restraints in the file 'talos.aco'.

Since this is not a calculation suited for the MPI scheduler, start CYANA first, then call the 'CALC.cya' macro from the prompt.

Hint: change to a cshell before running cyana (since talos needs a cshell to run):

csh

Automated NOESY assignment and structure calculation

We will perform an automated NOE restraint assignment and structure calculation by torsion angle dynamics.

The 'flya.prot' file from the automated resonance assignment will be used together with the (unassigned) NOESY peak lists to assign the NOESY peaks and to generate distance restraints. The structure is calculated in cycles, essentially testing the NOE assignment and iteratively refining it, in order to compute the three-dimensional structure of the protein.

Exercise 5: Run noeassign

Copy the 'flyabb' directory and give it the name 'noebb', then delete all the files and data we do not need to reduce clutter and have better oversight.

cp -r flyabb noebb
cd noebb
rm *asn.peaks *exp.peaks *.out *.job
rm -rf details

From the directory 'acoPREP' copy the calculated talos restraints ('talos.aco').

Inside the 'noebb' directory, use a text editor to edit the 'CALC.cya' file for noeassign as outlined.

The noeassign CALC macro

peaks:= cnoesy.peaks,nnoesy.peaks,aro.peaks 	
prot:= flya.prot                   		
restraints:= talos.aco                    		
tolerance:= 0.040,0.030,0.45            		
structures := 100,20                      		
steps:= 10000                       		
randomseed:= 434726    
                  		
noeassign peaks=$peaks prot=$prot autoaco

To speed up the calculation, you can set optionally in 'CALC.cya':

structures      :=50,10
steps=5000

These commands tell the program to calculate, in each cycle, 50 conformers, and to analyze the best 10 of them. 5000 torsion angle dynamics steps will be applied per conformer. If you do not set these option 100 conformers will be calculate, and the 20 best will be analyzed and kept.


When you are done preparing the macros as outlined run the calculation.

The automated NOE assignment and structure calculation will be performed by running the 'CALC.cya' macro:

cyana -n 33 CALC.cya

Doing this, basically means each processor will calculate 100/33=3 conformers. If you changed the setup to calculate 50 structures, you would start the calculation with 'cyana -n 25 CALC.cya'. 7 cycles of calculations will be preformed.

Statistics on the NOE assignment and the structure calculation will be in the file 'Table', which can also be produced with the command 'cyanatable -lp'.

The final structure will be 'final.pdb'. You can visualize it, for example, with the command

chimera final.pdb

The optimal residue range for superposition can be found with the command

cyana overlay final.pdb

Run noeassign with your 'CALC.cya' macro.

You can check the statistics (and success of 'noeassign') by running:

cyanatable

Creating the ligand library file for CYANA

In the next three exercises you will create the ligand library file for CYANA from scratch. Do this carefully and check your result, otherwise your structure calculation will not work as intended.

Exercise 6: Drawing the molecule and obtaining the SMILES code

  • make a copy of the libex and work in there (libexbb)
cp -r libex libexbb
cd libexbb

Go to the ZINC website.

Click on the Structure tab and draw the molecule using the supplied drawing (LIG.png) of the compound as a guide. Copy the SMILES code.


Hint: To look at the supplied image file in the terminal, use:

xdg-open Lig.pdf

Exercise 7: Converting the SMILES code to mol2

  • work in the copy of the libex directory ('cd libexbb')

There are many options and programs to do this, we outline two:


If you can use Avogadro (best):

For Mac OS download Avogadro from: Avogadro

Build -- > Insert --> SMILES

Paste the SMILES code

Extensions -- > Optimize Geometry

Save as

--> LIG.mol2 (*.mol2)


If you have to use chimera:

(If you are on the linux server chimera is installed)

Tools --> Structure Editing --> Build Structure Start Structure

--> SMILES string

set the Residue name to LIG (capital letters)

--> Apply Save your mol2 file as: LIG.mol2

Now, there is one issue we have to take care of: The intermolecular NOE assignments have to match the ligand structure assignment, otherwise the intermolecular NOEs will be wrong.

Using the text editor of your choice, manually change the "UNL1" in your mol2 file to "LIG".

Then open the supplied demoLIG.pdb structure in chimera, as well as the created mol2 structure.

chimera demoLIG.pdb LIG.mol2

First check the geometry (especially the rings and the stereochemistry). If it is wrong fix it!

Using the text editor of your choice, or more conveniently using chimera change the proton names in your mol2 file to match those of the pdb.

Hint: Overlay the two ligand structures in chimera.

Favorites --> Command Line

In the command line enter:

match #1 #0

Depending on how the models are loaded you may need to change the #? numbers. To see the model number use the Favorites --> Model Panel.

If chimera complains about "Unequal numbers of atoms chosen for evaluation", delete the pseudo atoms of 'demoLIG.pdb' temporarily for the overlay. In the command line

sel: @Q @Q? @Q??
delete sel

To rename selected atoms (control click) in the command line:

setattr a name HX sel

Hovering over atoms will display their names!

Exercise 8: Converting the mol2 file to a lib file for CYANA

  • work in the copy of the libex directory ('cd libexbb')
  • unpack the tool to convert the mol2 to a *.lib file
tar zxf cylib-2.0.tgz

run cylib with the options -nc -sc

./cylib-2.0/cylib -nc -sc LIG.mol2

this will create the LIG.lib file.

The -sc option keeps the angles of the rings fixed. We can do this since they are in this molecule either aromatic or have sp3 conjugated carbons in them, fixing the ring geometry. If they had to be flexible, you would need to keep the angeles flexible and supply additional restraints to close the rings.

To test the lib file we need CYANA:

Create a sequence file containing 'LIG 333' and name the file 'LIG.seq'.

Start CYANA This will read the CYANA library file correctlly but give you the error:

*** ERROR: Illegal residue name "LIG".
*** ERROR: Cannot read line 1:
                     LIG     333

Because we do not have an init file and have not read the 'LIG.lib' file yet, the program just tries to read the default sequence file in the directory, but the ligand is not yet in the library, so it fails...

read lib LIG.lib append
read seq LIG.seq

anneal

atoms select "* - &DUMMY"

pseudo=1
write pdb test.pdb selected

the command pseudo=1 ensures that the pseudo atoms will be in the written pdb file, 'atoms select "* - &DUMMY"' followed by 'write *.pdb selected' prevents the dummy atoms of the linker to be written to pdb.


Hint: Since you might have to do this a few times, until the library is working and correct, it might be worthwhile to create a 'init.cya' and a 'CALC.cya' macro with the respective commands. This to speed things up and prevent the error output shown above.


Carefully analyze the WARNING and ERROR messages if any.

Then take a look at your lig.pdb in chimera and check that the chemistry and bonds are all as expected (ring closure!)

chimera test.pdb

Again overlay the 'LIG.pdb' with the provided 'demoLIG.pdb'.

If there are any issues "go back to the drawing board" to fix the issues. Carefully check the names also of the pseudo atom names, since they are used in intermolecular-NOEs later.

To help find problems, you may use the command:

write lib LIG.lib names

This will write the library file containing actual atom names rather than numbers.

Alternative Exercise 6-8: Converting a pdb file to a lib file for CYANA

In case you were unsuccessful with exercises 6-8 in terms of getting a working ligand library file, do not dispair! There is an easy workaround that you may be able to use in the real case as well, converting a pdb file to a library file for CYANA.

Use Avogadro:

File --> Open

Open the 5c5aLig.pdb

Save as

LIG.mol2 (*.mol2)

Rename the Residue to LIG in the LIG.mol2 file.

 ./cylib-2.0/cylib -nc -sc LIG.mol2

Done!

You can run the tests outlined above, using anneal etc to test your library file.

Calculating the structure of the protein-ligand complex

Exercise 9: (Semi-automatic) Intermolecular cross peaks assignment and structure calculation

Since the molecular system contains protein and ligand, CYANA has to read the 'LIG.lib' file in addition to the regular 'cyana.lib' file. The sequence file needs to contain the protein and the ligand (and a linker to connect the two).

Copy the noebb directory and give it the name noecc, then delete all the previous, unnecessary output files to reduce clutter and have better oversight.

cp -r noebb noecc
cd noecc
rm *cycle* *.out *.job final* rama*

Update the 'init.cya' file in order to read the ligand library file and the sequence file containing the linker and the ligand.

Ad the 'read lib LIG.lib append' following the 'cyanalib' read command but before reading the sequence. 'append' is necessary, otherwise the 'cyana.lib' file will be overwritten by the 'LIG.lib' file.

rmsdrange:=15-111,333
cyanalib 

read lib LIG.lib append
read seq demoLong.seq

Intermolecular cross peaks we assign by supplying noeassign an intermolecular xeasy peak list with just the ligand resonances assigned. The ligand resonance were assigned manually and determined from an additional set of experiments (the semi-automatic part). Thereby the resonance assignment matches the ligand atom assignment in the library file created in the previous exercise.

The protein side will then be assigned by noeassign.

Update the your previous 'CALC.cya' macro by adding the intermol-NOEs.peaks to the peaks list and adding the keep=all option to the noeassign command:

peaks:= cnoesy.peaks,nnoesy.peaks,aro.peaks,intermol-NOEs.peaks 	
write_peaks_names=.true.
assign_noartifact:="** list=intermol-NOEs.peaks"
noeassign peaks=$peaks prot=$prot keep=all selectcombine="* - @LIG" autoaco

The command 'assign_noartifact' effectivly disables network anchoring tests for the ligand. Since the list supplied is cleaned and presumed artifact free, we are allowed to do this. We therby encourage the use of the intermolecular NEOs even if the support by other nearby NOEs is weak. The command 'write_peaks_names=.true.' ensures that the assigned peak list are written to file with the actual resonance names (this is not xeasy standard).

You can run the calculation again, commenting out (#) the 'assign_noartifact' command, and see the effect on the final structure.

'selectcombine' calls for testing for errors to be done different: Intermolecular peaks do not have to compete with intra protein peaks.


RUN this task using a different version of CYANA:

cyana -n 33 CALC.cya

Comparing the calculated NMR structure to an XRAY reference structure

Exercise 10: Compare the NMR structure to the Xray structure

Download (www.rcsb.org) the xray structure with ID: 5c5a

Use either a web-browser or the terminal:

wget 'https://files.rcsb.org/download/5c5a.pdb'

Using chimera it is possible to compare two structures, by overlaying and inspecting visually.

When you have your xray structure ready, load your calculated nmr structure and the xray structure in chimera.

Use to chimera specific commands to overlay the two structures and compare the structures visually.

Exercise 11: Preparing an xray structure to use within CYANA

Deposited structures often lack specific features. i.e. Xray structures usually lack proton coordinates.

Copy your noecc results to a new directory call regulabb, then delete all the previous, unnecessary output files to reduce clutter and have better oversight.

cp -r noecc regulabb
cd regulabb
rm *cycle* *.out *.job

After reading the sequence file, the pdb file can be read with the option unknown=warn or unknown=skip, this will then skip the parts of the molecule not specified in the sequence file.

read pdb xxxx.pdb unknown=warn

Other options to read pdb's:

read 5c5a.pdb unknown=warn hetatm new

where the option 'hetatm' allows for reading of coordinate labeled HETATM, rather than ATOM in the pdb. 'new' will read the sequence from the pdb.

To write back out pdb's and sequences:

write pdb XXX.pdb
write seq XXX.seq  

Inspect the pdb using chimera: Now, there are several issues besides HETATM, that make the comparison to the calculated NMR structure not possible within CYANA before you fix them. You may use a graphical text editor to fix them. In the end, you need to have a conformer of the complex ready to compare with the calculated NMR structure.

Best would be to practice the use of the 'regularize' command as well. This is however not really necessary in this particular case, since this xray structure contains proton coordinates. Using the regularize command one can get a structure calculated within CYANA that has these features but still is very close to the input structure of your choice.

Copy your 'LIG.lib' file and name it 'NUT.lib', in the 'NUT.lib' file change the residue name from LIG to NUT. The 'NUT.lib' file is necessary to read the original xray structure with ligand into CYANA.

Copy the 'demoLong.seq' file and name it 'demoLongEd.seq', in the 'demoLongEd.seq' file delete the linker residues.

Create an 'init.cya' macro with:

cyanalib
read lib LIG.lib append

Then create a 'CALC_reg.cya' macro with:

read lib NUT.lib append
read 5c5a.pdb unknown=warn hetatm new
write 5c5a_Ed.seq 
write 5c5a_Ed.pdb

#renumber and rename the ligand from 201 333, NUT to LIG
library rename "@NUT" residue=LIG
atoms select @LIG
atoms set residue=333

write 5c5a_renum.seq
write 5c5a_renum.pdb

#sequence with ligand but without linker
read demoLongEd.seq
read 5c5a_renum.pdb rigid unknown=warn

write XrayAChainRenum.pdb

initialize

read seq demoLong.seq
read pdb XrayAChainRenum.pdb unknown=warn

write pdb test.pdb

read pdb test.pdb
regularize steps=20000 link=LL keep

Execute the 'CALC_reg.cya' macro in the CYANA shell (or use only one processor, do not distribute the job):

cyana CALC_reg.cya

Exercise 12: Calclulate the RMSD of NMR vs. xray structure using a CYANA macro

Using the INCLAN language of CYANA (Writing and using INCLAN macros,Using INCLAN variables,Using INCLAN control statements) it is possible to write complex macros that interact with the FORTRAN code of CYANA. Reading internal variables and manipulating them to achieves custom task.

  • save the manually edited xray structure (exercise 11) or the the regularized xray structure (containing the ligand and called 'regula.pdb') as 'reg_xray.pdb' to use the macro below (or change the name in the macro accordingly).
  • what do you think about the RMSD, does the value make sense? Does the range make sense?

Below you find the commands for a macro (call it 'CALC_RMSD.cya') that will read the regularized xray structure and the calculated nmr structure, then calculating the rmsd of both the protein and ligand parts of the complex:

read demoLong.seq

rmsd range=15-111 structure=final.pdb reference=reg_xray.pdb

atom select "BACKBONE 15-111"
t=rmsdmean
j=rindex('333')
n=0
s=0.0
do i ifira(j) ifira(j+1)-1
  if (element(i).gt.1) then
    n=n+1
    s=s+displacement(i)
   end if
end do
print "RMSD of the LIG: ${s/n} ($n atoms)"

read pdb final.pdb
structure mean
write pdb mean.pdb

read pdb mean.pdb
read pdb reg_xray.pdb append

atom select "BACKBONE 15-111"
t=rmsdmean
atom select "WITHCOORDALL"
j=rindex('333')
n=0
s=0.0
do i ifira(j) ifira(j+1)-1
  if (element(i).gt.1.and.asel(i)) then
    n=n+1
    s=s+displacement(i)*2
  end if
end do
print "Displacement of the LIG (to ref xray): ${s/n} ($n atoms)"

Beyond The Basics: Improving the final structure

FLYA options

There are a variety of commands to modify FLYA runs to accommodate experimental labeling schemes or apply previous assignments etc...

Modify the chemical shift statistics used for assignment

Supply user-defined chemical shift statistics instead of standard BMRB statistics from library and replace the general statistics from 'cyana.lib' (CSTABLE).

  • average value and stddev from input chemical shift list 'shiftx.prot'
  • 'assigncs_sd:=bmrb' to use stddev from BMRB ('cyana.lib') instead of input chemical shift list
  • 'assigncs_sdfactor:=0.5' to scale BMRB stddev by given factor
shiftassign_statistics:=predicted.prot


Modify the reported statistics

Groups of atoms for which assignment statistics will be calculated and reported in the 'flya.txt' output file can be defined as:

analyzeassign_group := BB: N H CA CB C

In this case, the command defines a group called BB (a name that can be chosen freely) comprising the atoms N, H, CA, CB, C.


The optional parameter 'shiftreference=manREF.prot' specifies reference chemical shift list, used only for comparison in flya.tab, flya.txt, flya.pdf:

 shiftassign_reference:=manREF.prot

The same parameter may also be set as part of the flya command:

flya runs=10 assignpeaks=$peaks shiftreference=manREF.prot


Modify the expected peak lists

Specific labeling can be handled and peak list-specific atom selections can be applied.

To restrict the generation of expected peaks to a subset of atoms, here the backbone atoms:

command select_atoms
  atom select "N H CA CB C"
end


Input structures may be used to generate expected peaks for through-space experiments:

  • specify with parameter 'structure' of the command 'flya'
  • if parameter 'structure' is absent, a set of random structures is generated automatically
  • if set to blank ('structure='), no random structures are generated (if not needed because only through-bond spectra are used)
flya runs=10 assignpeaks=$peaks structure=XXX.pdb

Experimental peaks may also be employed as expected peak lists:

  • command N15NOESY_expect, reading input peak list N15NOESY_in.peaks
N15NOESY_expect :=N15NOESY_in


Keeping previously determined assignments

To keep input peak assignments in user peak assignments:

  • (partially) assigned input peak list XXX.peaks
  • parameter 'keepassigned' for 'loadspectra.cya'
loadspectra_keepassigned:=.true.

To fix input chemical shift assignments contained in a prot file

To do this i.e for backbone atoms extracted from the manREF.prot list:

Make a list of only the reference backbone chemical shifts by entering the CYANA commands:

read manREF.prot
  atom set "* - H N CA CB C" shift=none
write fix.prot

The file 'fix.prot' will contain the reference chemical shifts only for the backbone (and CB) atoms H, N, CA, CB, C'. Now you can repeat the assignment calculation by inserting the 'shiftassign_fix:=fix.prot' statement in 'CALC.cya' and choosing only the input peak lists that are relevant for sidechain assignment:

shiftassign_fix:=fix.prot


Chemical shift assignment using exclusively NOESY

  • increased population size with 'shiftassign_population=200'
  • see Schmidt et al. J. Biomol. NMR 57, 193-204 (2013)


Speeding up FLYA runs

Serves the fast automated chemical shift assignment and means the results in general are less accurate since either the populations are smaller, there are less parallel runs or the optimization schedule is modified.

In production runs, better results can be expected (at the expense of longer computation times) if these parameters are not set.


There are three parameters of the assignment algorithm that can be set in order to speed up the calculation.

Fixed number of generations in evolutionary optimization:

shiftassign_population=25

The population size for the genetic algorithm, i.e. how many assignments form one generation (25; chosen smaller than in normal production runs in order to speed up the calculation).

There is also an option to choose the "quick" optimization schedule:

shiftassign_quick=.true.

And last the 'runs' option can be set for flya as we did in the exercise ('flya runs=10').


neoassign options

To learn more about noeassign consult the tutorial Structure calculation with automated NOESY assignment. Other options for neoassign are described here: CYANA_Macro:_noeassign

Exercise 13: Mapping restraints onto a known structure

One can map the calculated restraints, such as distance restraints (upl/lol) onto a known structure (in the example here an xray structure). This is another approach to analyze restraints and their influence on the results.

Below you find the commands to accomplish this. You see by studying the commands, which files are needed to execute the macro. Therefore, create a new directory ('mkdir') or copy a directory containing the respective files. Delete what you do not need. Use the regularized xray structure from exercise 11.

Commands preceded by hashtags (#) are commented out, remove the hashtags if you want to use them. If you decide to use the intermo-NOEx-cycle7.peaks file, make sure to comment any commands you no longer need.

You need an init file:

rmsdrange:=15-111,333
cyanalib
read lib LIG.lib append

And the main macro (name it 'CALC_xraymap.cya'):

read seq demoLong.seq

The following block of commands, takes the assigned intermol.peaks list and calculates distance restraints from the peak intensities:

#peaks:=intermol-NOEs-cycle7.peaks
#calibration peaks=$peaks
#peaks calibrate simple
#write upl intermol.upl

The following block of commands, reads the 'final.upl' list (in this case of neoassign) and selects the intermolecular NOEs to LIG and writes them to file:

read upl final.upl
distance select "*, @LIG" info=full
write intermol.upl
read intermol.upl unknown=warn

#read upl lig.upl append
#read lol lig.lol

read regula.pdb unknown=warn

weight_vdw=0
overview intermol_xray.ovw
  • If the restraints do not match with the xray structure, does it mean they are wrong?
  • If you tried the two options, what is (are) the difference(s)?
  • Did you look at the LIG.upl/lol files in the demo_data folder, what are they? What type of NMR experiments are there to obtain them?

Exercise 14: Work on improving the final structure

Using what you have learned so far, employing some of the options of FLYA and noeassign, consider if it is possible to improve the resolution of the final structure.


General questions to answer regarding this task:

  • Are there additional restraints (or inputs) you could use for running FLYA or noeassign?
  • Are there other ways to calculate NMR structures within CYANA? Basic structure calculation starting from given restraints.
  • Are there additional NMR experiments you could measure, that are not supplied with the demo_data?