FREQUENTLY ASKED QUESTIONS


  -1- What services does R.E.D. Server Development provide?  
  -2- Who is authorized to use R.E.D. Server Development?  
  -3- How to use R.E.D. Server Development?  
  -4- How to create your own R.E.D. Server Development?  
  -5- How to get help about the PyRED program and R.E.D. Server Development?  
  -6- What files are needed to submit a job to R.E.D. Server Development?
  -7- How to download/access a project generated by R.E.D. Server Development?  
  -8- How are jobs handled by R.E.D. Server Development?  
  -9- How does a R.E.D. Server Development user know when her/his job starts and ends?  
-10- Can a R.E.D. Server Development user kill one of her/his job?  
-11- How long are the projects generated by R.E.D. Server Development kept available for downloading?  
-12- Which browser can be used to access R.E.D. Server Development?  
-13- What is the Internet address to download a R.E.D. Server Development/PyRED job?  
-14- What is the format of the Gaussian outputs generated by R.E.D. Server Development?  
-15- How to use NWChem, Q-Chem, Turbomole or Jaguar geometry optimization outputs in R.E.D. Server Development?  
-16- What are these error messages generated by R.E.D. Server Development when uploading input files?  
-17- Is there any limitation in R.E.D. Server Development and why?  
-18- How to generate a standalone graphical interface to parse R.E.D. Server Development outputs?  
-19- How to construct a PDB input file for R.E.D. Server Development using Jmol?  
-20- How to construct a PDB input file for R.E.D. Server Development using JSmol?  
-21- How to exchange data between users using R.E.D. Server Development?  
-22- How and when to contact the authors of R.E.D. Server Development?  
-23- How to cite R.E.D. Server Development?  



  -1- What services does R.E.D. Server Development provide?  

      -1.1- R.E.D. Server Development provides the software and hardware required for RESP and ESP atomic charge derivation, force field library building and force field parameter generation for new molecule(s) and/or new molecular fragment(s) (see Scheme 1):


Scheme 1

      a) Software
- In agreement with the Gaussian Inc. company, R.E.D. Server Development is authorized to interface the Gaussian program. Different versions of the Gaussian program are available (versions 2003 E.01, 2009 A.02, 2009 C.01, 2009 D.01, 2009 E.01 as well as the latest one 2016 C.01).
- In agreement with the GAMESS developers, the GAMESS-US (Sept 2018 R3) and Firefly (8.2.0) programs are also interfaced by R.E.D. Server Development.
- The latest version of the PyRED program developed by the q4md-forcefieldtools team is executed by R.E.D. Server Development.
      b) Hardware
R.E.D. Server Development runs on a computer cluster composed of E5430 quadri-core Intel® Xeon® processors: each job uses 8 cores (i. e. 2 Xeon cpus) and 16 or 64 GB RAM. A user can also request to use more powerful nodes (up to two ten-cores Xeon E5-2690 v2 processors (i. e. 20 cores) and up to 256 GB RAM used in each PyRED job; see FAQ 17 for more information).

      -1.2- Help for R.E.D. Server Development users is provided (see FAQ 5 for more information).

      -1.3- A tutorial is available.

      -1.4- A demonstration is also available from the Demo service at the R.E.D. Server Development home page.

      -1.5- For each project computed by R.E.D. Server Development a Jmol based graphical interface (in the php format) is generated allowing the user parsing and displaying the corresponding data (see also FAQ 18). Screen snapshots describing the graphical interface corresponding to the W-46 R.E.DD.B. project are described in Figure 1(A-B). The graphical interface as well as the corresponding data generated by R.E.D. Server Development for the F-45 and F-50 R.E.DD.B. projects are available here (with the last features incorporated in PyRED the "F-45" and "F-50" R.E.DD.B. projects can now be generated in a single R.E.D. job).


Figure 1(A-B)

      -1.6- For each project computed by R.E.D. Server Development a specific web address is created allowing users to directly access and download the corresponding data (see FAQ 7). A graphical interface to the qstat command is also available to follow the jobs in the R.E.D. Server Development queuing system (see FAQ 9).

      -1.7- Jmol applets (Javascript and Java versions) have been developed allowing any user building PDB input files for R.E.D. Server Development (see FAQ 19).

      -1.8- Data generated using R.E.D. Server Development can be automatically submitted to R.E.DD.B. if a R.E.D. Server Development user wants to use this option (see R.E.DD.B. FAQ 5 for more information).

      -1.9- Finally, the latest developments in term of force field development carried out by the q4md-forcefieldtools team will be provided through R.E.D. Server Development. The release of these new developments will be carried out from time to time, and the description of those new features released in R.E.D. Server Development will be available at the PyRED news web page.


  -2- Who is authorized to use R.E.D. Server Development?  

      All researchers (from the academic world and from private laboratories) can use R.E.D. Server Development.

      -2.1- In agreement with the developers of the GAMESS-US and Firefly programs any user is authorized to interface GAMESS-US and Firefly with R.E.D. Server Development.

      -2.2- In agreement with Gaussian Inc. only academic users are authorized to interface Gaussian with R.E.D. Server Development.


  -3- How to use R.E.D. Server Development?  

      -3.1- Registration to use R.E.D. Server Development is not mandatory. However, in agreements with the developers of the GAMESS-US, Firefly and Gaussian programs, the direct use of R.E.D. Server Development (i. e. without any registration step) is limited to the execution of GAMESS-US or Firefly.

      -3.2- R.E.D. Server Development provides also the opportunity to a user to create at any time a private space of work or account. In these conditions, a user has to Register at the R.E.D. Server Development home page and select "Registration to get a private account". To fulfill the agreements established between the developers of R.E.D. Server Development and these of the GAMESS-US, Firefly and Gaussian programs, it is requested to any researcher interested in registering as a R.E.D. Server Development user to certify if she/he is an academic users or not. However, after registration a user retains her/his account only if she/he keeps using it. Thus, after three months of inactivity the private account is automatically deleted, and a user has to re-register if she/he wants to interact with R.E.D. Server Development by using a private account again.

      The two ways to access R.E.D. Server Development are described in Table 1:

Ways to access
R.E.D. Server Dev.
  Registration
required
  Email of the
user required
  QM programs
interfaced (*)
  Account
used
Direct use of
R.E.D. Server Dev.
  No     No   GAMESS-US
Firefly
  A common public account
Use of a
private account
  Yes   Yes   Gaussian 2016
Gaussian 2009
Gaussian 2003
GAMESS-US
Firefly
  A private account per user
(*) During the Demo, any of the three QM programs can be interfaced.
Table 1

      -3.3- A private account created by a user is common to R.E.D. Server and R.E.D. Server Development.


  -4- How to create your own R.E.D. Server Development?  

      Any researcher can create her/his own R.E.D. Server Development. The source code of R.E.D. Server Development is distributed under the GNU General Public License, and is available for downloading for this purpose. Users interested in this option can contact us at contact_at_q4md-forcefieldtools.org. They should also contact the developers of GAMESS-US, Firefly and Gaussian if they decide to interface these programs.


  -5- How to get help about the PyRED program and R.E.D. Server Development?

      Any person interested in RESP or ESP charge derivation, force field library building and force field parameter generation for new molecule(s) and/or molecular fragment(s) can obtain help from the q4md-forcefieldtools team.

      -5.1- Each person having a private account and registered as a R.E.D. Server Development user can obtain a private assistance by using the Assist service provided at the R.E.D. Server Development home page. In this case, the questions asked by each R.E.D. Server Development user will remain hidden to the scientific community and the q4md-forcefieldtools team will try to provide specific answers to each R.E.D. Server Development user's problem.

      -5.2- For more general questions, a general public help is provided by using the q4md-forcefieldtools mailing list. Any researcher can participate in this mailing list by answering and/or sending queries to q4md-fft@q4md-forcefieldtools.org after registration at sympa@q4md-forcefieldtools.org. To register in the q4md-fft mailing list just send an email to sympa@q4md-forcefieldtools.org with "subscribe q4md-fft" in the email subject or body (to un-subscribe just send "unsubscribe q4md-fft"). Archives of the q4md-fft mailing list are public.
      When describing the problem you encountered with R.E.D. Server Development, please also provide the 'PXXXX' R.E.D. Server Development job name in the body of your email so that we can more easily assist you.

      -5.3- We are also registered in the AMBER and CCL mailing lists, and we will answer queries about the q4md-forcefield tools in these two mailing lists as well.


  -6- What files are needed to submit a job to R.E.D. Server Development?

      A R.E.D. Server Development user needs to prepare PDB input files for R.E.D. Server Development/PyRED (see FAQ 19). A description of the PDB format used is available in the readme.txt file (see the documentation). More generally this documentation describes the different files, which have to be uploaded and used as inputs. The input files to be provided depends on the type of computation to be performed, on the molecules involved in the job as well as on the modes of execution of the PyRED program. These input files have to be uploaded in R.E.D. Server Development as a single archive file. The definition of a PyRED job, as well as the upload of the archive file is performed from the submission service available at the R.E.D. Server Development home page:

      -6.1- The archive file has first to contain a series of Mol_red'n'.pdb input file(s) ('n' starts at 1, and is the number of the consecutive molecules involved in a PyRED job).
      It might also contain a series of geometry optimization Mol_red'n'.log file(s), as well as a Project.config, a System.config and/or a frcmod.user files.
      If Re_Fit is set to "On" in the "System.config" file a PyRED job has to be provided in the archive.

      -6.2- See also the tutorials.


  -7- How to download/access a project generated by R.E.D. Server Development?  

      There are two different cases depending on how the user interacts with R.E.D. Server Development:

      -7.1- When a researcher directly uses R.E.D. Server Development without any private account, the address of download for each project generated by R.E.D. Server Development is provided at the end of the input submission procedure (i. e. before the job effectively starts on R.E.D. Server Development).

      -7.2- When a researcher uses her/his private account, the project generated using R.E.D. Server Development can be downloaded via the Download service available at the R.E.D. Server Development home page. The download link is also provided at the end of the input submission procedure.


  -8- How are jobs handled by R.E.D. Server Development?  

      R.E.D. Server Development jobs are managed by Open PBS-Pro (version 18.1.3). A maximum of three or nine jobs per private or public account are allowed on R.E.D. Server Development, respectively. Other jobs are sent to the PBS queuing system. These simple rules insure that every user gets an equal access the computational resources provided by R.E.D. Server Development. This also means that the public account of R.E.D. Server Development can be quickly saturated if different researchers simultaneously use this account. More generally, the queuing system of R.E.D. Server Development will be adapted if it gets saturated (i. e. more nodes will be opened).


  -9- How does a R.E.D. Server Development user know when her/his job starts and ends?  

      There are two different cases depending on how the user interacts with R.E.D. Server Development:

      -9.1- When a researcher directly uses R.E.D. Server Development without any private account, she/he has the possibility to provide her/his email address at the beginning of the submission procedure. Thus, R.E.D. Server Development is able to reach the user to transmit information about the status of the job (initiation and termination). If a wrong email address is provided by error or if a user refuses to provide an email address, the user can still submit a job to R.E.D. Server Development and download the corresponding data. Indeed, the address of download (read FAQ 13 for more information about its format) is given to the user at the end of the submission procedure before the corresponding job terminates (or even starts). In the later case, as the submission procedure cannot guess when a R.E.D. Server Development job ends and cannot contact the user, the status of the job is not given to the user. The time taken by a PyRED job is difficult to estimate, in particular when geometry optimization is performed and/or when complex charge derivation involving multiple molecules and multiple conformations is requested. Consequently, a user has to look at the graphical interface of the qstat command to find out when a job executed using the public account terminates.

      -9.2- During the registration procedure to get a private account a researcher provides her/his email address. Thus, when a researcher uses a private account, she/he will receive three different emails from R.E.D. Server Development. The first one is sent when the inputs have been successfully uploaded to R.E.D. Server Development (the address of download of the data generated by R.E.D. Server Development is given to the user at the end of the submission procedure). The second one is sent by the R.E.D. Server Development queuing system when the job effectively starts. The third one is sent when the job has ended. At any time, a researcher can use the graphical interface of the qstat command to check the status of a R.E.D. Server Development job.

      Moreover, for each R.E.D. Server Development job the different steps carried out by the PyRED program are written in a log file. This file can be displayed by the owner of the job from the graphical interface of the qstat command. This allows any user checking the status of a given job at any time during its execution. The log file of a R.E.D. Server Development job executed using the public account remains public. On the contrary, a user must provide the login and password of her/his private account to display the .log file of a R.E.D. Server Development job executed using this account.


  -10- Can a R.E.D. Server Development user kill one of her/his job?  

      When using the Job deletion service available from the graphical interface of the qstat command a user can only kill the R.E.D. Server Development job(s) executed with her/his private account. Indeed, a R.E.D. Server Development job executed using the public account cannot be killed because this account is common and can be used anonymously. To kill a job a user has to provide the R.E.D. Server Development JOB NAME (obtained from the qstat graphical interface) with her/his login/password.

      The data corresponding to a killed job can now be downloaded by the user from her/his private account.


  -11- How long are the projects generated by R.E.D. Server Development kept available for downloading?  

      There are two different cases depending on how the user interacts with R.E.D. Server Development:

      -11.1- When a researcher directly uses R.E.D. Server Development without any private account, the public account of R.E.D. Server Development is used. Considering that this account can be extensively used and that tremendous amounts of data can be generated, R.E.D. Server Development projects generated using this account are kept only 5 days on the cluster hard drives.

      -11.2- On the contrary, when a researcher uses her/his private account, projects generated using R.E.D. Server Development are kept 20 days on the cluster hard drives.

      After these periods of time, the data are automatically deleted without warning.


  -12- Which browser can be used to access R.E.D. Server Development?  

      R.E.D. Server Development is not browser specific. The following browsers have been tested:
      - Mozilla/Firefox,
      - Chrome,
      - MacOS/Safari,
      - Microsoft/Edge,
      - Opera,
      - Pale Moon,
      - Brave,
      Extensive tests have been carried out with Firefox, Pale Moon, Brave and Chrome under Linux. Consequently, it is strongly recommended to use one of these browsers...

      Since the R.E.D. Server Development web site uses PHP sessions with cookies (to control the demo, assitance, submission and download procedures) and the javascript language (to control the shape of each window and to provide pieces of information in popup windows), cookies and javascript have to be enabled in the user browser.

      Java and javascript Jmol based applets are also used to visualize charge values and structures (read FAQ 19 and FAQ 20 for more information).


  -13- What is the Internet address to download a R.E.D. Server Development/PyRED job?  

      The Internet address/link to download a R.E.D. Server Development/PyRED job has the following format (it contains an embedded login and password): https://login:password@address_of_the_cluster/address_of_the_job_in_the_cluster. It is important to underline that the login and the password are asked twice to the user during the first authentication procedure: first by the web server, and then by the cluster. If registered in the navigator the login/passwd are then asked only once the next times.

      The particular format of this Internet link is not recognized by the old Internet Explorer navigator (tested versions: 6.x, 7.x and 9.0). To avoid this Internet explorer limitation the login and password have to be removed from the Internet link (i. e. one should use instead the following address: https://address-of-the-cluster/address-of-the-job), and provide the required login/password when demanded, or use another browser such as Firefox, Opera, Chrome, Safari or Edge.


  -14- What is the format of the Gaussian outputs generated by R.E.D. Server Development?

      In agreement with Gaussian Inc., the outputs generated by the Gaussian program are filtered by the PyRED program. Thus, only the pieces of information required in RESP and ESP charge derivation and force field library building are conserved in the Gaussian program outputs. If a user wants to get the complete Gaussian output, she/he will have to execute the Gaussian program without using R.E.D. Server Development on her/his own machine. In this case, downloading and using the R.E.D. tools available from the R.E.D. web site in a standalone mode is an alternative.


  -15- How to use NWChem, Q-Chem, Turbomole or Jaguar geometry optimization outputs in R.E.D. Server Development?

      Before anything else, it is recommended to check the latest updates available at the R.E.D. Server Development news. Indeed, this feature might have been released by the q4md-forcefieldtools team at the time a R.E.D. Server Development user needs it. If not available, optimized cartesian coordinates using NWChem, Q-Chem, Turbomole, Jaguar or ADF (among many others) have to be manually re-formatted into a format recognized by the R.E.D. program, i. e. a Gaussian or a GAMESS/Firefly like format. Read the R.E.D.-II FAQ about how to perform this task, and execute R.E.D. Server Development in "Mode 2".


  -16- What are these error messages generated by R.E.D. Server Development when uploading input files?

      When uploading the archive file the following error message may be obtained:

Warning*: ssh2_scp_send() [function.ssh2-scp-send ".../REDServer-Development/function.ssh2-scp-send"]: Unable to read source file in *~/REDServer-Development/xxxxxx.php* on line *xxx*
Upload of the archive file... [ Failed ]


      R.E.D. Server Development presents temporary internal network problems and the archive file cannot be uploaded. Simply wait for a couple of hours and try again. You might also try reducing the size of the uploaded QM file(s): read the R.E.D.-II FAQ about how to perform this task. Do not hesitate to contact the q4md-forcefieldtools team for reporting this problem if this problem lasts.


  -17- Is there any limitation in R.E.D. Server Development and why?  

      R.E.D. Server Development has been developed to generate empirical force fields for small molecules and/or molecular fragments. It is not designed to run computation on large molecular systems such as a polypeptide, a polynucleotide or a polysaccharide. Consequently, the maximum number of atoms for a molecule/conformation in an input file is set to 250 atoms. However, this does not mean R.E.D. Server Development cannot be used to generate a force field for a larger molecular system. In the later case a user only needs to split her/his "big" molecule/polymer into elementary building blocks with fully defined conformation(s) and use specific constraints during the charge fitting step to reconstruct the initial/big molecule. See the Tutorials related to the q4md-forcefield tools for more information.

      The maximum time (wall time) used by the queuing system for a R.E.D. Server Developement job is set to three weeks (or 21 days * 8 cores = 168 core days; see FAQ 1 and FAQ 8 for more information).

      A maximum of three or nine jobs per private or public account are allowed on R.E.D. Server Development, respectively (see FAQ 8 for more information). However, the queuing system of R.E.D. Server Development will be adapted if it gets saturated (i. e. more or less computational nodes/jobs per account will be made available).

      The maximum size of the upload archive file is set to 90 MB (see FAQ 6).

      Data generated by users are conserved five (public account) or twenty days (private account) on the R.E.D. Server Developement hard drives (see FAQ 11).

      By default a R.E.D. Server Development job is executed on a node with two quad-cores Intel® Xeon® E5430 processor (i. e. 8 cores) and 64 GB RAM. Such computer harware is generally sufficient for quantum mechanics (QM) computation involving small molecules. However for a big molecular system and/or when using a large QM basis set the computational time dramatically increases. Thus to limit this time problem a user can request to work on a more powerful node, i. e. to use up to two ten-cores IntelĀ® Xeon E5-2690 v2 processor (i. e. 20 cores) and up to 256 GB RAM.

      A user can contact the q4md-forcefieldtools team at contact_at_q4md-forcefieldtools.org to request to work with a powerful node and/or to bypass one of these limits.


  -18- How to generate a standalone graphical interface to parse R.E.D. Server Development outputs?  

      For any job generated by R.E.D. Server Development, a user can generate a Jmol based graphical interface to parse R.E.D. Server Development outputs. In this case download the following csh script here, read the header of the script to understand how to use it, and execute it by using "csh Web-InterfacePyRED.csh PXXX".


  -19- How to construct a PDB input file for R.E.D. Server Development using Jmol?  

      Jmol based java applets have been developed allowing any user building PDB input files for R.E.D. Server Development. Six templates are available for building any type of nucleosides, amino acids, monosaccharides, inorganic complexes and organic or heterocyclic molecules from an input molecule. To be able to use this feature, a user has to install the J2SE Runtime Environment (JRE) on her/his front-end workstation (see help here about using the Pale Moon browser and configuring java on Linux).

      It is important to underline that Firefox and Google no longer supports the NPAPI technology, that is required for Java applets. Starting with Chrome version 42 and Firefox 52 java plugins cannot be enabled in the browsers. However one can use the Pale Moon browser instead of Firefox/Chrome with java based Jmol applets, or Firefox or Chrome with javascript based JSmol applets (see FAQ 20 for more information about JSmol).


  -20- How to construct a PDB input file for R.E.D. Server Development using JSmol?  

      JSmol based javascript applets have been developed allowing any user building PDB input files for R.E.D. Server Development. Six templates are available for building any type of nucleosides, amino acids, monosaccharides, inorganic complexes and organic or heterocyclic molecules from an input molecule. This applet combines the six previous ones, and allows users loading her/his input model to modify it.

      The PyRED program cannot yet add missing hydrogens to a structure, and these hydrogen atoms are required in quantum chemistry computation. However, one can use a JSmol applet to perform this task (i) open one the javascript applets provided above, and create a new structure from an input molecule, (ii) if some hydrogen atoms are missing in the construction open the Jmol console (right-click and select "Console"), (iii) add the missing hydrogens [write in the console: "calculate hydrogens"; in case of a bad topology remove/add the additional/missing hydrogens/connectivities using the Jmol Model kit (right-click and select "Computation/Model kit")], and (iv) save the new set of Cartesian coordinates in the PDB file format (write in the console: "write coord PDB"; see the Jmol documentation). This new PDB file can now be used as input for a R.E.D. Server Development/PyRED job.

      A new JSME/JSmol applet is now available: the q4md-fft builder combines all the features present in the former applets. One can load a three-dimensional structure from a database using its name or load a model to modify it, or create a two-dimensional sketch and convert it into a three-dimensional structure or use a simplified molecular input line entry specification (smiles) and far more.


  -21- How to exchange data between users using R.E.D. Server Development?  

      For each R.E.D. Server Development job a specific web address is created allowing users to download/directly access the data generated. When using the public account these web links can be exchanged between users to study and compare data. On the contrary, we strongly discourage users to exchange web links generated for their private account since a private password has been generated for this account.


  -22- How and when to contact the authors of R.E.D. Server Development?

      -21.1- Ask your questions about the PyRED program and R.E.D. Server Development in the help services available at the q4md-forcefieldtools web site (see FAQ 5 for more information).

      -21.2- If you find a bug, if you have a request or if you need a new feature to be implemented in R.E.D. Server Development, contact the q4md-forcefieldtools team. We will always try to help.


  -23- How to cite R.E.D. Server Development?  

E. Vanquelef, S. Simon, G. Marquant, E. Garcia, G. Klimerak, J. C. Delepine, P. Cieplak & F.-Y. Dupradeau, R.E.D. Server: a web service for deriving RESP and ESP charges and building force field libraries for new molecules and molecular fragments, Nucl. Acids Res. (Web server issue), 2011, 39, W511-W517.

See here.




Last update of this web page: September 16th, 2024.

Université de Picardie Jules Verne. Sanford Burnham Prebys Medical Discovery Institute.
© 2009-2024. All rights reserved.

Valid XHTML 1.0 Strict CSS Valide !