refinement of two different filaments

[Laurence serre]

Dear Eman2 developers

version: Eman 2.99.47

I used subtomogram averaging method to solve a model which consists in two species of filaments (A and B-filament). I would like to determine (if possible) the relative polarity of one with respect to the other. The two filaments cannot be refined together as they have different periodicity.

1.     -I picked up the subtomograms using Eman2’s filament option and extract the particles

2.     -I solved and orientated the thichker filament (A-filament) with “ e2spt_refine_new.py” and obtain a good model at 9.0 A resolution using a known model of A-filament.

e2spt_refine_new.py --ptcls=sets/particles.lst --ref=A-filament.hdf --startres=50.0 --goldstandard --sym=c1 --iters=p3,t2,p,t,r,d --keep=0.95 --maxres=0.0 --minres=0.0 --parallel=thread:22 --threads=20 --curve

 

3.     -To solve the B-filament, I used ptcls=spt_01/aliptcls3d_06.lst as input and a mask defined around B-filament location and a known model of B-filament. Final model is about 9.0 A resolution.

e2spt_refine_new.py --ptcls=spt_01/aliptcls3d_06.lst --ref=B-filament.hdf --startres=50.0 --goldstandard --sym=c1 --iters=p3,t2,p,t,r,d --keep=0.95 --loadali2d=spt_01/aliptcls2d_06.lst --maxres=0.0 --minres=0.0 --parallel=thread:28 --threads=24 --curve --mask=mask-B-filament.hdf

 

4.     -I checked the polarity of the final refined B-filament and it always matches the polarity of the input reference (I used both polarities for B-filament).

Therefore, it seems that the initial orientation parameters from spt_01/aliptcls3d_06.lst are not considered to orient the B-filament.

 

My question is: Is there a way to prevent the flipping of the aligned particles of the A-filament ? 

…and to determine whether there is or there isn’t a specific polarity of the B-filament relatively to the A-filament ?

 

I am not sure this is really clear…sorry. Thanks in advance for your help.

Laurence Serre, Grenoble Institute of Neuosciences, FRANCE

[Steve Ludtke]

EMAN2 has moved away from "stable versions" to rolling releases with distribution via conda-forge. Your version is ~4 years old, so there is a good chance that the problem has been fixed. Regardless, the first thing to try is to get a current release with the current installation strategy and see if the problem still occurs.

https://blake.bcm.edu/emanwiki/doku.php?id=eman2:install:condainstall

Note that there is still a choice between installing the last compiled binary or installing from source, which gives the flexibility of picking any specific GitHub time-point. Binaries are generated only when one of the devs thinks they have done something important. Minor changes and bugfixes hang around in github until that time.

Actually, it looks like it's been a while, so maybe time to push a new binary to catch all of the smaller changes over the last few months.

--
--
----------------------------------------------------------------------------------------------
You received this message because you are subscribed to the Google
Groups "EMAN2" group.
To post to this group, send email to em...@googlegroups.com
To unsubscribe from this group, send email to eman2+un...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/eman2

---
You received this message because you are subscribed to the Google Groups "EMAN2" group.
To unsubscribe from this group and stop receiving emails from it, send an email to eman2+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/eman2/03e5fbb2-4037-4fed-9407-1f5177721896n%40googlegroups.com.

[Muyuan Chen]

If you want to preserve the alignment of spt_01/aliptcls3d_06.lst, use --localrefine instead of --curve. 

You may find some useful information in the tutorial. 

https://blake.bcm.edu/emanwiki/doku.php?id=eman2:e2tomo_p22

It is for viruses but it includes discussion on how to refine different parts of the same protein with different symmetry. 

If the A filament refinement looks good, and you can see some density of B filament from the A filament structure, maybe it is possible to get the B filament structure from the particles themselves, similar to the virus tail finding, so it avoids the bias from the reference. But if you have to use a reference, and you are using --localrefine, make sure to move the reference to the roughly B filament location in the A filament average before the refinement. Otherwise it would just align the A filament density to the B filament reference. 

If you want to determine if there is a specific polarity, consider adding --breaksym=d1 in addition to --localrefine. But for that you will need to re-extract particles with shifted coordinates first so the B filament is roughly along the z axis. 

Muyuan

[Laurence serre]

thanks a lot for your suggestions..

L.

[laurence]

Dear Steven

I cannot update the eman version as our machine is too old. By mistake I erased the eman2.99.47 which was running finely. Is threre a way to download the old version and the protocol to reinstall it on our computer as everything as migrated on wiki now?

thanks a lot for your help.

I feel a bit desperate.

L.

Laurence Serre

Grenoble Institut Neurosciences

France

[Steve Ludtke]

So, not quite sure what "as our machine is too old" means in this context. While the binaries do have some minimal CPU version requirements, compiling from source should be possible on pretty much any linux machine (is it linux?).   Can you provide some information about the computer (exact processor, operating system & version) we might offer an alternative solution. Yes, we should still be able to find the old binaries for you if you really need them. 

To view this discussion visit https://groups.google.com/d/msgid/eman2/ab59e601-7ef1-4631-925d-3db91871b789n%40googlegroups.com.

[laurence]

Dear Steve

yes maybe saying that our machine is too old, is a bit strong. here are the information about our Linux computer. thanks a lot for your optimistic reply.

NAME="CentOS Linux"
VERSION="7 (Core)"
ID="centos"
ID_LIKE="rhel fedora"
VERSION_ID="7"
PRETTY_NAME="CentOS Linux 7 (Core)"

Architecture:          x86_64
CPU op-mode(s):        32-bit, 64-bit
Byte Order:            Little Endian
CPU(s):                64
On-line CPU(s) list:   0-63
Thread(s) per core:    2
Core(s) per socket:    32
Socket(s):             1
NUMA node(s):          1
Vendor ID:             AuthenticAMD
CPU family:            23
Model:                 49
Model name:            AMD EPYC 7502P 32-Core Processor
Stepping:              0
CPU MHz:               1500.000
CPU max MHz:           2500.0000
CPU min MHz:           1500.0000
BogoMIPS:              4999.92
Virtualization:        AMD-V
L1d cache:             32K
L1i cache:             32K
L2 cache:              512K
L3 cache:              16384K
NUMA node0 CPU(s):     0-63

Laurence Serre

Grenoble Institut of Neurosciences

France

[Steve Ludtke]

Ahh, I see. The problem is not the computer. The computer is absolutely fine. The problem is running CentOS 7, which came out in 2014, and reached end of life (no security patches) in 2024. 

The problem with CentOS isn't strictly it's age, but that they made some very strange choices with the default C compiler they used for it which make it incompatible with a lot of software. The solution for this was (and still is) to install an alternative "newer" C++ compiler on the machine. As you stated, at one time we also made binaries which would be compatible with this for people who couldn't handle the compiler installation.  With CentOS being retired now, getting some of the older packages may be a little more challenging if you don't have them installed.

If you do have alternative (newer) C++ compilers on the machine via alternatives or some other mechanism, then following the source compilation instructions:

https://blake.bcm.edu/emanwiki/doku.php?id=eman2:install:sourceinstall

should work fine.

If you really want to install the now very out of date 2.99.47 binaries, this should work again:

https://cryoem.bcm.edu/downloads/view_eman2_version/33

To view this discussion visit https://groups.google.com/d/msgid/eman2/f8100933-a55a-45ff-9fd3-9a1f0183852bn%40googlegroups.com.

[laurence]

thanks Steve

indeed  I tried to install the eman2 last version, the GCC version was a problem. (our current GCC version is 4.8-something and version 5 was required).

I will try to install a new compiler otherwise we will carry on with the oldest (but fine) 2.99.47 version until we have another computer with a compatible OS.

L.