The Zhang Lab Message Boards

Hi,

I obtained 5 models for sequences with or without a mutant, respectively. There are 4 models which can be classified in 2 groups with a similar structure. Whether should I chose the higher c-score in 2 groups or just pick the two Top 1 models to run RMSD?

Hi everyone,

I need help setting up slurm script for I-Tasser for running it at cluster. I have attached script with this post, any help greatly appreciated. thanks.

Hi,

I setup a C-QUARK run on Saturday and it still says "The sequence has been submitted and in processing..." I'm assuming that the job is still in the queue and has not started. Do you know what is the current wait times before a job starts in C-QUARK? Or is there an error with the submission? Any help would be much appreciated. Thanks. Jeff

I am new for using QUARK online and have a simple question regarding the result. In the QUARK results I received, there are top 5 final structure models in 3D. Are these models shown in the order of probability?

I would appreciate your reply.

Best regards,
Koichi Kato

Hello all,

I'm not sure if this question corresponds in here, but I will give it a try.
I have been trying to run FUpred locally, but I'm receiving the following error,

pwd: /media/Disco4T/jcaceres/FUpred
cat: /media/Disco4T/jcaceres/FUpred: Is a directory
ERROR! Cannot locate /media/Disco4T/jcaceres/fupred_data/uniclust30_2018_08_a3m_db for
-hhblitsdb=/media/Disco4T/jcaceres/fupred_data/uniclust30_2018_08
cp: cannot stat 'protein.aln': No such file or directory
cp: cannot stat 'protein.a3m': No such file or directory
cp: cannot stat 'deepmsa_protein.a3m': No such file or directory
ERROR: cannot open /tmp/jZP3SB7Lgh/YIbhTz02zC.in.a3m!
cp: cannot stat 'psitmp.mtx': No such file or directory
cp: cannot stat 'psitmp.pssm': No such file or directory
Illegal division by zero at ./run-FUpred.pl line 238.

I have downloaded the uniclust30db, which is a folder that includes the file uniclust30_2018_08_a3m_db. I have tried to explicitly give uniclust30_2018_08_a3m_db to the hhblitsdb variable, but still not working. Can please someone help me?

Many thanks,
Javier.

my I-TASSER results for job ID S52695 Shows
C- Score: nan
estimated TM and RMSD score also shows nan for all predicted 5 models.
could you please help solve this problem.

Thank you

I have a protein that has been predicted to be 1 domain using standalone ThreaDom, and I wish to use I-TASSER standalone to predict the structure of it. Is it possible to do this?

Dear Prof. Zhang,
since the EC-prediction is a structure based method, in my opinion when the c-score and TM-score for structure prediction do not meet the threshold values, -1.5 and 0.5 respectively, then the structure prediction and hence the EC assignment should not be accepted. Could you kindly confirm if I am thinking correctly ?
Thanks,
ambarishnag.

Dear Developer,

I obtained my I-TASSER results. I observed that the top 5 modelspresenetd C-sore between -5 and -4. How could I selet the best model? Moreover,the values are not so encouraging. Thus: how could I handle these models? They are unrialable? Could I improveone the selected one?

Thank you in advance.

Hi,

I have recently registered in I-Tasser server with my academic email id. I have email and password, but i am unable to download the I-Tasser standalone software.

Kindly, help me out for this issue.

Regards,
Anurag

Hi,
if the COACH-predicted EC-score value is < 1.1 then should the EC assignment be discarded ?
Also, say for a given sequence, the c-score is < -1.5 and TM-score < 0.5, but EC-score > 1.1. In this case, should be function assignment be accepted ?
Thanks,
ambarishnag.

Dear Zhang Lab Team,

I submitted my protein.pdb for refinement in ModRefine. I received the link to my results.

The problem is I cannot open the link in my browser. Is there any other way I can get my results?

Kind regards,
Jennifer

Hello. I submitted two jobs earlier this week, one to I-TASSER and one to I-TASSER-MR. I checked on the I-TASSER-MR job a few minutes ago (job T1159ntZlp) and it says it's still in que (job_waiting). I've also not received an email about the I-TASSER job, which leaves me to believe its also still in que. Is there something I've inputted incorrectly with these jobs, since they seem to be in que and not running? Thank you.

Dear I-TASSER Team,

I am a frequent user of the I-TASSER Tool. I am studying this particular protein that I have submitted for modelling reference ID S527124.

Initially I modelled with a restraint by providing a pdb file as template. The template similarity was low and therefore the model was not so good.

I have tried submitting the sequence without specific restraint so that the tool can use 10 different templates for modelling hoping I get a better model.
Whenever I submit I keep getting the same model. Even when I have not specified any option nor provided a template.

Please assist me to be able to submit without restraint and get a better model.

Kind regards,
Jennifer.

Hello!

I submitted a job a week ago, I was wondering how long does it take before the job starts? I'm guessing there is a long queue at the moment, hence, the long wait?

Thanks

Hi,
Could someone tell me what mean a C-score nan.
A model was predicted in I-TASSER but the results were:

C-score= nan
Estimated TM-score = nan±-nan
Estimated RMSD = nan±-nanÅ

Thanks.

Hello All ,

I have been interested by BSpred as the tool output would be really useful for a major part of the prepossessing step on a drug discovery project I am working on. I downloaded the local version as I have several thousands proteins to process. However, following the instructions in the Readme and running "runBSpred.pl" with my "data directory" as command line argument, the program executes with multiples error messages and generate an empty prediction file. I attempted to run "runBSpred.pl" using the example directory in the package as command line argument and the error persists. I have attached a screenshot of the error messages.

My own attempts to debug and fix the Perl code have been unsuccessful. it is my understanding though that the first error occurs after a call to an executable called "solve"(located in INSTALL_BSpred//library/bin) at line 122 of "exp.pl"(located in INSTALL_BSpred/bin) , "exp.pl" was itself called at line 250 of "BSpredmod"(located in INSTALL_BSpred/bin). "BSpredmod" appears to be the main perl file after "runBSpred.pl" initialize some variables to retrieve seq.txt from the "data directory".

Would it be possible to fix the issue and upload an updated downloadable package on the BSpred website in a certain timeline ? Besides, I also believe there is an issue with the webserver too as i never received a response after submitting an amino acid sequence as a test case.

Dear all,

I would like to predict the secondary structure using previously generated MSA.

However, I looked up the PSSPred perl script, it seems that it requires not only a MSA generated from PSI-BLAST, but also the .chk file. Does anybody know if I can somehow generate this file having my own MSA?

Regards,
Suzanne

I received my STRUM results 2/21/20 and have not been able to access them due to receiving this 403 FORBIDDEN error when I click on the link. Any advice on what to do?

Dear developers,
I wondered whether the job I submitted on 25th Feb 2020 (job ID: S523627) is still running or is it stuck for some technical reason?
Thank you in advance

Hi,
I am running runI-TASSER.pl on a cluster using SLURM with the following options

#SBATCH --nodes=1
#SBATCH --ntasks-per-node=16
#SBATCH --error itasser.err

-runstyle gnuparallel
-LBS true
-EC true
-GO false

The code runs to completion and I get the desired output files. However, I also get the following errors in the itasser.err file:

At line 447 of file ppa1.f
Fortran runtime error: Bad real number in item 4 of list input
At line 447 of file ppa1.f
Fortran runtime error: Bad real number in item 4 of list input
At line 1074 of file zal33.f
Fortran runtime error: Bad real number in item 4 of list input
Exception in thread "main" java.lang.NumberFormatException: For input string: "-NAN."
at sun.misc.FloatingDecimal.readJavaFormatString(FloatingDecimal.java:2043)
at sun.misc.FloatingDecimal.parseFloat(FloatingDecimal.java:122)
at java.lang.Float.parseFloat(Float.java:451)
at java.lang.Float.valueOf(Float.java:416)
at c.a(c.java)
at c.main(c.java)
Exception in thread "main" java.lang.NumberFormatException: For input string: "-NAN."
at sun.misc.FloatingDecimal.readJavaFormatString(FloatingDecimal.java:2043)
at sun.misc.FloatingDecimal.parseFloat(FloatingDecimal.java:122)
at java.lang.Float.parseFloat(Float.java:451)
at java.lang.Float.valueOf(Float.java:416)
at e.a(e.java)
at e.main(e.java)
FORTRAN STOP
Useless use of /d modifier in transliteration operator at /scratch/anag/JSD_qury/parse_blast.pl line 314.
Useless use of /d modifier in transliteration operator at ./parse_blast.pl line 314.
7456.80user 824.86system 1:16:35elapsed 180%CPU (0avgtext+0avgdata 3512956maxresident)k
24513344inputs+791672outputs (2major+161024267minor)pagefaults 0swaps

Is this something that I should be concerned about ? And is there a way to fix this issue ?

Thanks,
ambarishnag.

dear zhanglab:
i have been submit both of two files on 28 January,and 2 February by FG-MD and I-TASSER .and up to now ,i just get two result email by I-TASSER sever on 3 Feb and 5 Feb。ID is S518115 and S519188 .but i didn't receive any email about FG-MD , i have check the job on web is successful done. So i guess....is there something wrong with me? and unfortunately i don't remember the ID in FG-MD ，the file name is C1171S and wt. thanks for your help.i'm sincerely looking forward your reply.
your user

Does a slower computer just take longer to generate the output or can it make the output lower quality due to a "maximum time limit" of some type for simulations?

I am asking because I have noticed a general trend (not confirmed of course) of lower C-score structures from the I-TASSER distribution outputs when compared to the online server.

Hi,
Here are lines from a cscore file:

==================================================================
| Estimated accuracy of your models |
==================================================================
Estimated Clusters (N=20200)
--------------------- ------------------
Model# C-score TM-score RMSD (A) #decoys density
------ ------- ---------- -------- ------- -------
model1 -4.21 0.27+-0.08 14.1+-3.8 3333 0.021
model2 -4.31 2892 0.019
model3 -4.46 2244 0.016
model4 -4.75 1974 0.012
model5 -4.82 1625 0.011

The cscore values are less than -1.5 and the TM-score of the best model is 0.27
The first I-TASSER paper from 2008 mentions using a C-score cutoff > -1.5 for the models of correct topology and that TM-score > 0.5 corresponds approximately to two structures of the similar topology.
Given my best C-score is -4.21 and the corresponding TM-score is 0.27, what do these values signify about the best predicted structure ?
Thanks.

What is average C-I Tasser time to process the sequence?

I want to get the similarity score of two protein contact map.

Now, I have obtained protein contact map by ResPRE. According to the readme.txt in CEThreader, I know that I should use CEdecomposition to get .ce file for each protein first and use CEthreading to compare two .ce files to get final similarity score. But I have some questions:

Which option should I use in CEdecomposition ( template or query ) to get .ce file?
For template: how can I get .pdb and .mtx file?
For query: how can I get .meta and .mtx file?

Finally in CEthreading, what can I do to get .prf(query profile) file ?

Thank you!

Hi
I would to like to predict my proteins that have already 2D to tertiary structure_? I don't use the sequence that ı have the proteins. I ll submit the 2D structure to see them in 3D structure. Is that possible to apply by I-Tasser_? or you have ant recommendation. Thanks

18/5000

How should we predict protein configurations containing d-type amino acids，Like this sequence：CAGVYCCLISYPGADYKRITLK，The 12th proline is a d-type amino acid，How do I specify this residue？

Dear all,
If you can pls tell the computational power used for ITASSER in CASP ?
I shall be thankful to you!
My email is : mehtaroadies@gmail.com
Thanks!

Hello
I want to build a 3D model for ALSIN protein. When i insert my amino acid sequence it doesn't work because it is larger that 1500 residues. It suggests to add the domains separately and merge them together. how can i merge these models together? Is there a tool in I-TASSER which can merge these model domains together and give a single final model?