Re: [AMBER] Amber16 installation: Segmentation fault (core dumped) error when execute antechamber

From: Yuan Hu <Yuan.Hu.alkermes.com>
Date: Tue, 23 Jan 2018 06:41:28 +0000

Hi Dave,

Thanks for the reply. I will answer your questions in line.
I think there may be some interfering with anaconda3. Please see more details in question 4.

1. What is the result of typing "gcc --version"?

gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-16)

2. Do programs not from the antechamber directory work OK? What happens if you type "make test" in the $AMBERHOME directory? [You can hit control-C if all the tests appear to be failing...no need to wait for thousands of error messages....; on the other hand, it might be that non-antehcamber tests generally pass, and that would be good information to know.]

Yes, other programs work fine. I went through all the programs in the $AMBERHOME/bin, all the failed programs are from antechamber directory. Interestingly, several tools in that folder are fine, such as bondtype, database, espgen, match, parmcal.
Here is the "make test" result, the errors in the log file are attached to the end.

make[1]: Leaving directory `/usr/local/software/amber16/AmberTools/test'
1993 file comparisons passed
18 file comparisons failed
16 tests experienced errors
Test log file saved as /usr/local/software/amber16/logs/test_at_serial/2018-01-22_23-56-27.log
No test diffs to save!

3. Did you source the $AMBERHOME/amber.sh file in the terminal session that is failing?

Yes. I sourced $AMBERHOME/amber.sh

4. Do you have older versions of amber on your machine that might somehow be interfering?

I don't have older versions of AMBER on the machine, but I found the same antechamber and other AmberTool programs in /usr/local/software/anaconda3/bin/. The anaconda3 was installed after I built AMBER.
Something I want to mention, I installed AMBER16 first, and did a quick test of antechamber. it works fine right after the installation. Then I installed anaconda3 to install openmm, pdbfixer, lomap and yank. After those installations, I tested antechamber again, and fount it broken.
I tried to remove anacoda3 from my $PATH, $LD_LIBRARY_PATH, $PYTHONPATH, but got no luck to make antechamber work.
$echo $PATH
/usr/local/software/amber16/bin::/usr/local/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/usr/local/cuda-8.0/bin

$echo $LD_LIBRARY_PATH
/usr/local/software/amber16/lib:/usr/lib64:/usr/lib:/usr/local/lib:/lib:/usr/local/cuda-8.0/lib64/:/usr/lib64/openmpi/lib:/usr/local/cuda/lib64

$echo $PYTHONPATH
/usr/lib64/python2.7/site-packages:/usr/lib/python2.7/site-packages:/usr/local/software/amber16/lib/python2.7/site-packages

$antechamber
/usr/local/software/amber16/bin/antechamber: line 4: 7113 Segmentation fault (core dumped) $AMBERHOME/bin/to_be_dispatched/antechamber "$."


-Yuan






Errors in the "make test" log file:
make[2]: Entering directory `/usr/local/software/amber16/AmberTools/test/antechamber'
cd bondtype && ./Run.bondtype
  ./Run.bondtype: Program error
make[2]: [test] Error 1 (ignored)
cd c60 && ./Run.c60
  ./Run.c60: Program error
make[2]: [test] Error 1 (ignored)
cd ccif && ./Run.ccif
  ./Run.ccif: Program error
make[2]: [test] Error 1 (ignored)
cd charge_equlibration && ./Run.equ_charge
Step1: run antechamber to generate resp charges with '-eq 0'
  ./Run.equ_charge: Program error
  ./Run.equ_charge: The last command emitted:
/usr/local/software/amber16/bin/antechamber: line 4: 14157 Segmentation fault (core dumped) $AMBERHOME/bin/to_be_dispatched/antechamber "$."
make[2]: [test] Error 1 (ignored)
cd charmm && ./Run.charmm
  ./Run.charmm: Program error
make[2]: [test] Error 1 (ignored)
cd chemokine && ./Run.1b2t
  ./Run.1b2t: Program error
make[2]: [test] Error 1 (ignored)
cd tp && ./Run.tp
  ./Run.tp: Program error
make[2]: [test] Error 1 (ignored)
cd tp && ./Run.tp2
  ./Run.tp2: Program error
make[2]: [test] Error 1 (ignored)
cd translate && ./Run.2rh1
Check dimensions
Read in the second GPCR protein, 2RH1 and make check its dimensions.

 /usr/local/software/amber16/bin/translate -i 2rh1.pdb -f pdb -c check
/usr/local/software/amber16/bin/translate: line 4: 14216 Segmentation fault (core dumped) $AMBERHOME/bin/to_be_dispatched/translate "$."
Align to Z-axis

>From a graphic program, we can see that the 7TM helixes is parallel to Y-axis.
Now we want to align the molecule to make the helixes in Z-axis. To do so, we need to
specify two points so that after the alignment, the origin is the center of the two
and both points are in Z-axis. Here are the two points: x1 and x2 are -33.088 (the average x)
z1 and z2 are 7.0287 (the averge z) and y1 = -14.578 (the min y), y2 = 50.061 (the max y)
Of course, one may use pick up other points,for instance, using the coordinates of two atoms.
 /usr/local/software/amber16/bin/translate -i 2rh1.pdb -f pdb -c alignz -x1 -33.088 -x2 -33.088
-y1 -14.578 -y2 50.061 -z1 7.0287 -z2 7.0287 -o 2rh1_Z.pdb


/usr/local/software/amber16/bin/translate: line 4: 14225 Segmentation fault (core dumped) $AMBERHOME/bin/to_be_dispatched/translate "$."

Check dimensions again

 /usr/local/software/amber16/bin/translate -i 2rh1_Z.pdb -f pdb -c check
/usr/local/software/amber16/bin/translate: line 4: 14233 Segmentation fault (core dumped) $AMBERHOME/bin/to_be_dispatched/translate "$."


Now we want to rotate 60 degree the molecule along Z-axis. first select two points in Z-axis
Here are the two points: x1, x2, y1, y2 are 0, z1 = 10, z2 = -10
(of course you could use other numbers for z1 and z2)

 /usr/local/software/amber16/bin/translate -i 2rh1_Z.pdb -f pdb -c rotate2 -x1 0 -x2 0
-y1 0 -y2 0 -z1 10 -z2 -10 -o 2rh1_Z60.pdb -d 60

/usr/local/software/amber16/bin/translate: line 4: 14240 Segmentation fault (core dumped) $AMBERHOME/bin/to_be_dispatched/translate "$."

Check dimensions again

 /usr/local/software/amber16/bin/translate -i 2rh1_Z60.pdb -f pdb -c check
/usr/local/software/amber16/bin/translate: line 4: 14247 Segmentation fault (core dumped) $AMBERHOME/bin/to_be_dispatched/translate "$."


Now we want to rotate 90 degree the molecule along Z-axis. first select two points in Z-axis
Here are the two points: x1, x2, y1, y2 are 0, z1 = 10, z2 = -10
(of course you could use other numbers for z1 and z2)

 /usr/local/software/amber16/bin/translate -i 2rh1_Z.pdb -f pdb -c rotate2 -x1 0 -x2 0
-y1 0 -y2 0 -z1 10 -z2 -10 -o 2rh1_Z90.pdb -d 90

/usr/local/software/amber16/bin/translate: line 4: 14254 Segmentation fault (core dumped) $AMBERHOME/bin/to_be_dispatched/translate "$."

Check dimensions again

 /usr/local/software/amber16/bin/translate -i 2rh1_Z90.pdb -f pdb -c check
/usr/local/software/amber16/bin/translate: line 4: 14261 Segmentation fault (core dumped) $AMBERHOME/bin/to_be_dispatched/translate "$."


Now we want to rotate 180 degree the molecule along Z-axis. first select two points in Z-axis
Here are the two points: x1, x2, y1, y2 are 0, z1 = 10, z2 = -10
(of course you could use other numbers for z1 and z2)

 /usr/local/software/amber16/bin/translate -i 2rh1_Z.pdb -f pdb -c rotate2 -x1 0 -x2 0
-y1 0 -y2 0 -z1 10 -z2 -10 -o 2rh1_Z180.pdb -d 180

/usr/local/software/amber16/bin/translate: line 4: 14268 Segmentation fault (core dumped) $AMBERHOME/bin/to_be_dispatched/translate "$."

Check dimensions again

 /usr/local/software/amber16/bin/translate -i 2rh1_Z180.pdb -f pdb -c check
/usr/local/software/amber16/bin/translate: line 4: 14275 Segmentation fault (core dumped) $AMBERHOME/bin/to_be_dispatched/translate "$."


possible FAILURE: file 2rh1_Z.pdb does not exist.
==============================================================
possible FAILURE: file 2rh1_Z60.pdb does not exist.
==============================================================
possible FAILURE: file 2rh1_Z90.pdb does not exist.
==============================================================
possible FAILURE: file 2rh1_Z180.pdb does not exist.
==============================================================
cd ash && ./Run.ash
  ./Run.ash: Program error
make[2]: [test] Error 1 (ignored)
cd sustiva && ./Run.sustiva
  ./Run.sustiva: Program error
make[2]: [test] Error 1 (ignored)
cd fluorescein && ./Run.fluorescein
  ./Run.fluorescein: Program error
make[2]: [test] Error 1 (ignored)
cd guanine_amber && ./Run.guanine_amber
  ./Run.guanine_amber: Program error
make[2]: [test] Error 1 (ignored)
cd fpph && ./Run.fpph_resp
  ./Run.fpph_resp: Program error
make[2]: [test] Error 1 (ignored)
cd residuegen && ./Run.ala
  ./Run.ala: Program error
make[2]: [test] Error 1 (ignored)
cd parmchk2 && ./Run.parmchk2
/usr/local/software/amber16/bin/parmchk2: line 4: 14370 Segmentation fault (core dumped) $AMBERHOME/bin/to_be_dispatched/parmchk2 "$."
  ./Run.parmchk2: Program error
make[2]: [test] Error 1 (ignored)
cd acdoctor && ./Run.acdoctor

Here is a bad structure in pdb format, mpd.pdb. We want to use acdoctor
to find problems and antechamber to fix them. First of all, run acdoctor :
/usr/local/software/amber16/bin/acdoctor -i mpd.pdb -f pdb >& mpd_pdb.out

Since acdoctor indicates the molecule has more than one unit, it is
likely there are missing bonds. We convert the file from pdb to mol2
without judging bond types by using flag -j 5 :
/usr/local/software/amber16/bin/antechamber -i mpd.pdb -fi pdb -o mpd.mol2 -fo mol2 -j 5 -at sybyl -dr no

  ./Run.acdoctor: Program error
make[2]: [test] Error 1 (ignored)
make[2]: Leaving directory `/usr/local/software/amber16/AmberTools/test/antechamber'

This message may contain confidential or privileged information. If you are not the intended recipient, please advise the sender immediately by reply e-mail and delete this message and any attachments without retaining a copy. Thank you for your cooperation.
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Jan 22 2018 - 23:00:02 PST
Custom Search