Bug or feature (?) : residue numbering not understood

[persalteas]

Dear Dr Xiang-Jun,

About residue numberings. After annotating a structure, outputing in JSON, looking at the "nts" section, residues are organized by chain, and within a chain, the "index_chain" field starts at one (expected behavior, very good). However, in some cases, it starts at last chain's last index_chain plus one, is there any reason for that ?

My spotted case is 4w29 (a large ribosome), i am looking at chain BA, which starts with index_chain = 77, following chain AX's index_chain = 76.
I checked, the chains are not contiguous in space, nothing indicates they could be linked together. Other chains in the structure start at 1.

Execution output below (nothing special spotted), using DSSR v1.9.9:
$ x3dna-dssr -i=/home/persalteas/Data/RNA/3D/RNAcifs/4w29.cif --json --auxfile=no > 4w29.json

Code: [Select]

Processing file '/home/persalteas/Data/RNA/3D/RNAcifs/4w29.cif'
[i] '/home/persalteas/Data/RNA/3D/RNAcifs/4w29.cif' taken as in .cif format by file extension.
  1.AA.U.434 0.121
  1.AA.U.723 0.121
  1.AA.U.1126 0.135
  1.AA.U.1301 0.123
  1.AW.U.8 0.123
  1.BA.U.1939 0.122
  1.BA.U.2887 0.121
  1.CA.U.68Q 0.126
  1.CA.U.669 0.123
    total number of nucleotides: 9398
    total number of amino acids: 14040
[w] arginine [AD.ARG3] -- NH1/NH2 naming swapped
[w] arginine [AD.ARG118] -- NH1/NH2 naming swapped
[w] arginine [AT.ARG89] -- NH1/NH2 naming swapped
[w] arginine [AY.ARG351] -- NH1/NH2 naming swapped
[w] arginine [AY.ARG354] -- NH1/NH2 naming swapped
[w] arginine [AY.ARG357] -- NH1/NH2 naming swapped
[w] arginine [AY.ARG396] -- NH1/NH2 naming swapped
[w] arginine [AY.ARG484] -- NH1/NH2 naming swapped
[w] arginine [AY.ARG504] -- NH1/NH2 naming swapped
[w] Residue AV.G1 missing planar atom [ C8 ]
    total number of base pairs: 4601
    total number of multiplets: 620
    total number of helices: 263
    total number of stems: 601
    total number of isolated WC/wobble pairs: 151
    total number of atom-base capping interactions: 589
    total number of splayed-apart dinucleotides: 1238
                        consolidated into units: 797
    total number of hairpin loops: 228
    total number of bulges: 118
    total number of internal loops: 245
    total number of junctions: 116
    total number of non-loop single-stranded segments: 73
    total number of kissing loops: 13
    total number of A-minor (types I and II) motifs: 191
    total number of eXtended A-minor (type X) motifs: 219
    total number of ribose zippers: 94
    total number of kink turns: 29
[w]  number-of-residues=658 -- in shortened form
[w]  number-of-residues=658 -- in shortened form
[w]  cross-paired segments in separate chains, be *careful* with .dbn

Time used: 00:00:02:56
Thanks,

Louis

[xiangjun]

Hi Louis,

Thanks for using DSSR and for posting your questions on the 3DNA Forum. I will look into the reported "issue" and get back to you on the Forum, hopefully before long...

Best regards,

Xiang-Jun

[xiangjun]

Hi Louis,

The issue has been fixed and will be released in DSSR Pro. Due to a lack of funding support, this is the only way my effort on DSSR can be justified. I will ensure that paid users always receive top-notch support.

Xiang-Jun

[persalteas]

Hi,

Thank you for the quick fix !

I have an academic licence for DSSR2, but i can consider "DSSR Pro", what is the difference and how much does it cost ? (The CTV website displays a blank page)

Due to a lack of funding support, this is the only way my effort on DSSR can be justified. I will ensure that paid users always receive top-notch support.

I understand. While i entirely trust you since you have demonstrated efficiency and skills in developping DSSR and quickly replying to questions on this forum, the CTV is not very engaging, since the website regularly has issues. My first visit was in September 2020, blank page, I had to ask techventures<at>columbia.edu by email to fix it. And today i have the same issue. And i could read in the forum that other people faced problems.
I suggest you summarize the content of the page in a forum post, to provide a reliable source of information about DSSR licensing, maybe ?

[xiangjun]

Hi Louis,

Sorry for your difficult experience in accessing the CTV website for DSSR. I've communicated with the supporting staffs, and hopefully situation would improve as a result of your feedback.

Basically, the DSSR Pro academic license charges a one-time fee of $1000 per seat. It comes with a one-year support directly from me. I do not expect another major release of DSSR in the next 4 years. As a result, the annual cost averages $250, or about a tenth of the cost of a journal publication or a two-night hotel stay. Email and video meetings are included in the one-year support. Following your suggestion, I will create a post on DSSR Pro academic license in the "Site announcements" section. [Note added on May 31, 2021: I've added a post titled "Clarification on DSSR licensing".]

Best regards,

Xiang-Jun

Below in the copied from the CTV website on DSSR Pro:

License Name

DSSR-Pro Academic ($1000, 1 seat)

Description

This academic license is for one user and includes the DSSR Pro software (macOS, linux, and windows) along with the Pro Manual. It is for people working on open academic research at, or for, an academic institution or similar non-profit. Any commercial use requires purchase of a commercial license. It is a one-time fee and does not cover future major releases of the code. Minor releases with bug-fixes and small updates are included in the fee. Thank you for purchasing DSSR. Funds help to support further development of the DSSR software product.

DSSR Pro has more functionality than DSSR Basic, including: (i) homology modeling via in silico base mutations, (ii) easy generation of regular helical models, and (iii) creation of customized structures with user-specific base sequences and rigid-body parameters. DSSR Pro supersedes 3DNA completely while DSSR basic has no modeling capabilities and less analysis/annotation features. DSSR Pro integrates disparate analysis and modeling programs of 3DNA under one umbrella, and offers an easier to use interface. DSSR Pro includes an in-depth user manual, and one year technical support directly from the developer. DSSR basic is provided AS IS, and does not include any support.

Pricing Information

Total Price $1,000.00

[persalteas]

Hi Xiang-Jun,

Thanks for the copy-paste. I will discuss with my team, but yes, we may go for it.
If the CTV team wants more information on the problem, do not hesitate to ask me (i tried with different browsers, but nothing succeeds).

Louis

[xiangjun]

Hi Louis,

I will discuss with my team, but yes, we may go for it.

Thanks!

If the CTV team wants more information on the problem, do not hesitate to ask me (i tried with different browsers, but nothing succeeds).

Yes, please provide screenshots and post them here. I've contacted the CTV support team, and they will view this thread.

Best regards,

Xiang-Jun

[persalteas]

Dear Xiang-Jun, dear CTV team,

Attached to this post a screenshot of the page at URL http://innovation.columbia.edu/technologies/CU20391/ when i open it in Mozilla Firefox (tested on Linux and Windows 10) or Microsoft Edge (Windows 10).
Firefox version : 88.0.1 (but i had the same issue in past months with past versions)
Edge version : 91.0.864.37 (but same)


(Also notice the lack of HTTPS.)

Hope this helps,

Louis

[xiangjun]

Hi Louis,

Thank you for your input! The blank-page screenshot of the CTV DSSR website using the browsers/OSes you tested is surprising. It's possible that using HTTP rather than HTTPS is a problem. The 3DNA Forum, on the other hand, uses HTTP rather than HTTPS and apparently you can access it without difficulty.

I just tried using Firefox and discovered that the DSSR website on CTV was blocked by default, and I had to add columbia.edu to my trusted sites list to see the landing page. I've mostly used Safari and Chrome (on macOS) at work and haven't had any issues. As an example, see the screenshot from the CTV DSSR website. Because Chrome and Safari account for such a large percentage of browser usage, the CTV and many users may be unaware of the problem. Could you please try Chrome and Safari and report back your findings?

I'll inform the CTV team tomorrow (today is Memorial Day in the United States) about the access issue and the use of HTTP rather than HTTPs. Hopefully, things will turn around.

Best regards,

Xiang-Jun

[persalteas]

Hi,

The page works in Chromium and Chrome, works in Opera, and does not work in Firefox, Edge, or Vivaldi. I have no macOS computer to test Safari, sorry.

Louis

[xiangjun]

Hi Louis,

Thank you for letting us know which browsers work and which don't when it comes to accessing the CTV websites. I've forwarded your findings to the DSSR-affiliated CTV support team. I'm guessing the problem isn't specific to the DSSR website on CTV, but rather a problem that affects a variety of websites.

My spotted case is 4w29 (a large ribosome), i am looking at chain BA, which starts with index_chain = 77, following chain AX's index_chain = 76.
I checked, the chains are not contiguous in space, nothing indicates they could be linked together. Other chains in the structure start at 1.

Now back to your initial question. I think it is a bug in DSSR that is only triggered in a rare case like 4w29: AX is an RNA chain, however, it has VAL77 at the end, as shown below:

ATOM   60690  C  C8    . A   X  24 76   ? 55.433   -163.384 92.500  1.00 140.84 ?  76   A   AX C8    1
ATOM   60691  N  N7    . A   X  24 76   ? 55.656   -163.215 91.217  1.00 141.03 ?  76   A   AX N7    1
ATOM   60692  C  C5    . A   X  24 76   ? 54.538   -162.520 90.779  1.00 141.71 ?  76   A   AX C5    1
ATOM   60693  C  C6    . A   X  24 76   ? 54.160   -162.033 89.512  1.00 142.30 ?  76   A   AX C6    1
ATOM   60694  N  N6    . A   X  24 76   ? 54.901   -162.183 88.412  1.00 143.14 ?  76   A   AX N6    1
ATOM   60695  N  N1    . A   X  24 76   ? 52.983   -161.377 89.410  1.00 142.62 ?  76   A   AX N1    1
ATOM   60696  C  C2    . A   X  24 76   ? 52.240   -161.228 90.513  1.00 142.49 ?  76   A   AX C2    1
ATOM   60697  N  N3    . A   X  24 76   ? 52.490   -161.642 91.754  1.00 142.54 ?  76   A   AX N3    1
ATOM   60698  C  C4    . A   X  24 76   ? 53.665   -162.290 91.829  1.00 141.84 ?  76   A   AX C4    1
ATOM   60699  N  N     . VAL X  24 77   ? 50.042   -163.172 98.420  1.00 50.64  ?  77   VAL AX N     1
ATOM   60700  C  CA    . VAL X  24 77   ? 51.296   -162.409 98.368  1.00 47.13  ?  77   VAL AX CA    1
ATOM   60701  C  C     . VAL X  24 77   ? 52.559   -163.283 98.400  1.00 44.56  ?  77   VAL AX C     1
ATOM   60702  O  O     . VAL X  24 77   ? 52.567   -164.326 99.054  1.00 40.69  ?  77   VAL AX O     1
ATOM   60703  C  CB    . VAL X  24 77   ? 51.361   -161.302 99.446  1.00 46.79  ?  77   VAL AX CB    1
ATOM   60704  C  CG1   . VAL X  24 77   ? 50.129   -160.404 99.438  1.00 47.04  ?  77   VAL AX CG1   1
ATOM   60705  C  CG2   . VAL X  24 77   ? 51.655   -161.873 100.828 1.00 46.46  ?  77   VAL AX CG2   1

With DSSR v2.3.1-2021jun01 (to be released), the index_chain starts at 1 instead of 77 for chain BA. Is that what you'd expect? Please confirm.

Best regards,

Xiang-Jun

[persalteas]

Yes, this is what i expect.

[xiangjun]

Hi Louis,

Thank you for confirming. The CTV website now has the DSSR Pro v2.3.1-2021jun01 available. If you have previously downloaded a version, please update to the latest version. If you have any questions, please let me know.

I greatly appreciate your support of the DSSR project by purchasing a Pro license. I sincerely hope you will find the quality of the software and my support to be well worth the money you have invested.

Best regards,

Xiang-Jun