Validation of Consensus Run: Difference between revisions

From NESG Wiki
Jump to navigation Jump to search
(Created page with '== '''Validation of Consensus Run''' == === '''Verification of Consensus Assignments''' === # Create a working subdirectory, for example, <tt>cyana21/calc2</tt>. # Copy consens…')
 
No edit summary
 
(12 intermediate revisions by the same user not shown)
Line 1: Line 1:
== '''Validation of Consensus Run''' ==
== '''Verification of Consensus Assignments''' ==


=== '''Verification of Consensus Assignments''' ===
&nbsp;&nbsp;&nbsp; 1.&nbsp; Create a working subdirectory, for example, <tt>cyana21/calc2</tt>.


# Create a working subdirectory, for example, <tt>cyana21/calc2</tt>.
&nbsp;&nbsp;&nbsp; 2.&nbsp; Copy consensus peaklists (<tt>n_cons.peaks</tt>, <tt>ali_cons.peaks</tt> and <tt>aro_cons.peaks</tt>) into this subdirectory.  
# Copy consensus peaklists (<tt>n_cons.peaks</tt>, <tt>ali_cons.peaks</tt> and <tt>aro_cons.peaks</tt>) into this subdirectory.
** If using CARA, re-integrate the peakslist and export them.
# Copy <tt>init.cya</tt>, <tt>XXXX.seq</tt>, <tt>XXXX.prot</tt> and <tt>stereofound.cya</tt> from the CYANA automated run (<tt>cyana21/calc2</tt>). Rename <tt>stereofound.cya</tt> into <tt>stereo.cya</tt>
# Copy <tt>GridSearch.cya</tt>, <tt>CALC.cya</tt> and <tt>talos.aco</tt> from the <tt>found</tt> subdirectory. Modify <tt>GridSearch.cya</tt> to read <tt>cons.upl</tt> instead of <tt>short.upl</tt>.
# Create UPL file <tt>cons.upl</tt> using run_caliba.cya (see below).
# In CYANA 2.1 run <tt>GridSearch.cya</tt>
# In CYANA 2.1 run CALC.cya for manual structure calculation with only consensus constraints.
# Clean violations the same was is in [[NESG:UsingFOUNDakaHABAS|Verifying Local NOE Network with FOUND (HABAS)]] and try to keep TF < 1.0.
** For the violated NOE, re-integrate the peaks first if it is overintegrated (in XEASY).
** Unassign peaks, which appear at spectral artifacts.
** If violation persisit while peak integrals are correct, then modify the calibration parameters.
# Verify the additional stereospecific assignments made in this run. Swap the atom groups if needed.


You may want to run an additional iteration of consensus assignment, though it is usually not necessary:
&nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; If using CARA, re-integrate the peakslist and export them.


=== '''Acceptance Criteria''' ===
&nbsp;&nbsp;&nbsp; 3.&nbsp; Copy <tt>init.cya</tt>, <tt>XXXX.seq</tt>, <tt>XXXX.prot</tt> and <tt>stereofound.cya</tt> from the CYANA automated run (<tt>cyana21/calc2</tt>). Rename <tt>stereofound.cya</tt> into <tt>stereo.cya</tt>


A successful consensus run shall satisfy:
&nbsp;&nbsp;&nbsp; 4.&nbsp; Copy <tt>GridSearch.cya</tt>, <tt>CALC.cya</tt> and <tt>talos.aco</tt> from the <tt>found</tt> subdirectory. Modify <tt>GridSearch.cya</tt> to read <tt>cons.upl</tt> instead of <tt>short.upl</tt>.  
# Both program CYANA and AutoStructure automatic structure calculation deliver converged structure.
# The structures from program CYANA and AutoStructure are similar, the RMSD value between the mean structures from both programs should be less than the sum of pairwise RMSD values of the two structure families.
# The manual run of CYANA structure calculation with distance constraints from consensus assigned NOEs delivers converged structure, the pairwise RMSD of the structure family shall be less than 3A.
# The yield of consensus assigned NOEs shall be large than 50%.


=== '''Example scripts''' ===
&nbsp;&nbsp;&nbsp; 5.&nbsp; Create UPL file <tt>cons.upl</tt> using run_caliba.cya (see below).


==== '''run_caliba.cya''' ====
&nbsp;&nbsp;&nbsp; 6.&nbsp; In CYANA 2.1 run <tt>GridSearch.cya</tt>


<nowiki>
&nbsp;&nbsp;&nbsp; 7.&nbsp; In CYANA 2.1 run CALC.cya for manual structure calculation with only consensus constraints.
protocol:=caliba.log
 
&nbsp;&nbsp;&nbsp; 8.&nbsp; Clean violations the same was is in Verifying Local NOE Network with [[FOUND|FOUND (HABAS)]] and try to keep TF &lt; 1.0.
 
&nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; For the violated NOE, re-integrate the peaks first if it is overintegrated (in XEASY).
 
&nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; Unassign peaks, which appear at spectral artifacts.
 
&nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; If violation persisit while peak integrals are correct, then modify the calibration parameters.
 
&nbsp;&nbsp;&nbsp; 9.&nbsp; Verify the additional stereospecific assignments made in this run. Swap the atom groups if needed.
 
&nbsp;&nbsp;&nbsp; You may want to run an additional iteration of consensus assignment, though it is usually not necessary.<br>  
 
== '''Acceptance Criteria'''  ==
 
A successful consensus run shall satisfy:
 
#Both program CYANA and AutoStructure automatic structure calculation deliver converged structure.
#The structures from program CYANA and AutoStructure are similar, the RMSD value between the mean structures from both programs should be less than the sum of pairwise RMSD values of the two structure families.
#The manual run of CYANA structure calculation with distance constraints from consensus assigned NOEs delivers converged structure, the pairwise RMSD of the structure family shall be less than 3A.
#The yield of consensus assigned NOEs shall be large than 50%.
 
<br>
 
== '''Example scripts'''  ==
 
=== '''run_caliba.cya'''  ===
<pre>protocol:=caliba.log
read prot $name
read prot $name
   
   
Line 51: Line 62:
   
   
write upl cons.upl
write upl cons.upl
protocol:=
protocol:=</pre>  
</nowiki>
<br> <br><br>
 
%COMMENT%
 
 
-- Main.GaohuaLiu - 26 Jan 2007

Latest revision as of 21:15, 6 January 2010

Verification of Consensus Assignments

    1.  Create a working subdirectory, for example, cyana21/calc2.

    2.  Copy consensus peaklists (n_cons.peaks, ali_cons.peaks and aro_cons.peaks) into this subdirectory.

            If using CARA, re-integrate the peakslist and export them.

    3.  Copy init.cya, XXXX.seq, XXXX.prot and stereofound.cya from the CYANA automated run (cyana21/calc2). Rename stereofound.cya into stereo.cya

    4.  Copy GridSearch.cya, CALC.cya and talos.aco from the found subdirectory. Modify GridSearch.cya to read cons.upl instead of short.upl.

    5.  Create UPL file cons.upl using run_caliba.cya (see below).

    6.  In CYANA 2.1 run GridSearch.cya

    7.  In CYANA 2.1 run CALC.cya for manual structure calculation with only consensus constraints.

    8.  Clean violations the same was is in Verifying Local NOE Network with FOUND (HABAS) and try to keep TF < 1.0.

            For the violated NOE, re-integrate the peaks first if it is overintegrated (in XEASY).

            Unassign peaks, which appear at spectral artifacts.

            If violation persisit while peak integrals are correct, then modify the calibration parameters.

    9.  Verify the additional stereospecific assignments made in this run. Swap the atom groups if needed.

    You may want to run an additional iteration of consensus assignment, though it is usually not necessary.

Acceptance Criteria

A successful consensus run shall satisfy:

  1. Both program CYANA and AutoStructure automatic structure calculation deliver converged structure.
  2. The structures from program CYANA and AutoStructure are similar, the RMSD value between the mean structures from both programs should be less than the sum of pairwise RMSD values of the two structure families.
  3. The manual run of CYANA structure calculation with distance constraints from consensus assigned NOEs delivers converged structure, the pairwise RMSD of the structure family shall be less than 3A.
  4. The yield of consensus assigned NOEs shall be large than 50%.


Example scripts

run_caliba.cya

protocol:=caliba.log
read prot $name
 
read peaks n_cons
peaks select "**"
peaks set volume=abs(volume)
caliba bb=1.7E+06
 
read peaks ali_cons
peaks select "**"
peaks set volume=abs(volume)
caliba bb=1.7E+06
 
read peaks aro_cons
peaks select "**"
peaks set volume=abs(volume)
caliba bb=1.7E+06
 
write upl cons.upl
protocol:=