Since securing funding for the X3DNA-DSSR project through an NIH R24 grant, I have dedicated myself to continuously advancing and refining the software tool. DSSR is a robust and mature tool with a professional user manual. Getting it up and running is straightforward, and assistance for installation is rarely required. Common usages are also already addressed, allowing me to focus on developing new features and addressing edge cases proactively.
- I regularly test DSSR using the latest weekly updates from the Protein Data Bank (PDB), identifying and resolving issues before users report them.
- I actively monitor the 3DNA Forum, providing timely responses to user queries, addressing reported issues, and introducing new features when necessary.
- Writing papers and blog posts can be effective methods to highlight areas where clarification and improvement are needed.
- Collaborating with other researchers often leads to enhancements in DSSR.
- I monitor how DSSR is cited, respond quickly to any reported issues, and contact authors when necessary.
As an example, I noticed a recent article titled 'Structural Analysis of Uridine Modifications in Solved RNA Structures' (https://doi.org/10.1093/nargab/lqaf197) by Arteaga and Znosko, where DSSR was cited as below:
Secondary structure elements (SSEs) containing uridine modifications were identified and annotated using Dissecting the Spatial Structure of RNA (DSSR). Corresponding SSEs containing canonical uridine were identified via RNA Characterization of Secondary Structure Motifs (CoSSMos) and also annotated using DSSR.
Identification and characterization of SSEs RNA SSEs were identified and characterized using the software Disecting the Spatial Structure of RNA (DSSR) [76]. DSSR, an RNA-specific successor to 3DNA [77], was employed to analyze RNA structures containing U modifications. The analysis was performed using default parameters, and all output data were saved in a ∗.json file format. For each ∗.json file, relevant information regarding the U modification residues was extracted. This information included the type of SSE, its associated nucleotide sequence, hydrogen bond acceptor/donor groups, base stacking (π stacking) interactions, sugar pucker, and glycosidic angle. The distance cutoff for identifying hydrogen bonding and base stacking interactions was set to 4.0 Å, as per DSSR’s default settings.
The following citation draw my attention:
An additional 21 structures were excluded from the analysis due to missing atoms in the RCSB PDB entries, DSSR processing issues, incomplete SSEs, or the inability to clip the SSEs (Supplementary Table S2).
I am curious to understand the DSSR processing issues referred to in their study. Upon reviewing 'Supplemental Table S2. Structures excluded from the analysis,' I noted that the following 15 PDB entries were listed as 'Unable to be annotated by DSSR': 4U3M, 4U3U, 4U4Q, 4U4R, 4U4U, 4U52, 4V88, 4V9O, 4V9P, 5FL8, 5TBW, 6I7V, 6SV4, 6Z1P, and 7QVP.
These are all large RNA structures, with 13 of them containing more than 10,000 nucleotides each. I am unsure which version of DSSR was used in their study; however, when using the current version (v2.7.2-2026jan12), I can process these PDB entries without encountering any issues. In cases such as these, or any issues related to 3DNA/DSSR, users are encouraged to reach out. I always aim to address inquiries promptly.
References
- Arteaga SJ, Znosko BM. Structural analysis of uridine modifications in solved RNA structures. NAR Genomics and Bioinformatics. 2026;8:lqaf197. https://doi.org/10.1093/nargab/lqaf197.
