Re: IUCr 2026 keynote proposal on crystal data standards

Dear COMCIFS,

Please see information below regarding a proposed keynote for IUCr 2026. Prof Kurlin's keynote has been put forward by the Mathematical Crystallography Commission and is supported by CommDat, and he has requested COMCIFS support as well. I believe the subject matter is something that COMCIFS should support, and given the time constraints (by 1st March) I have taken the liberty of advising Prof Kurlin that we support the proposed keynote without consulting COMCIFS first. If there are any objections, please let me know.

best wishes,

James.

---------- Forwarded message ---------
From: Prof Vitaliy Kurlin <vitaliy.kurlin@gmail.com>
Date: Thu, 27 Feb 2025 at 09:54
Subject: IUCr 2026 keynote proposal on crystal data standards
To: <jamesrhester@gmail.com>

[edited]

------------------------------
Speaker. Vitaliy Kurlin, vitaliy.kurlin@liverpool.ac.uk
Materials Innovation Factory, University of Liverpool, UK

Area of Expertise. Data Science for applications in crystallography
(complete and continuous invariants of periodic crystals for reliable
data standards and ultra-fast detection of near duplicates in massive
datasets), materials chemistry (geographic-style maps for
configuration spaces of atomic structures), and structural biology
(invertible and bi-continuous invariants of protein backbones for
detecting duplicates in the PDB), see more details at
https://kurlin.org/research-papers.php#Geometric-Data-Science

Title. A mathematically justified data standard for crystal structures.

Abstract. Crystal structures are determined in a rigid form and hence
keep all their properties under rigid motion within the same ambient
environment. However, structures that have different rigid shapes can
substantially differ by properties and hence should be reliably
distinguished, for example polymorphs with different solubility.
Conventional representations based on reduced cells discontinuously
change under almost any perturbation of atoms, which led to the
accumulation of near-duplicates in major databases of experimental
structures [1]. This ambiguity was recently resolved by generically
complete and continuous invariants that distinguish all non-duplicate
periodic crystals in the Cambridge Structural Database within a few
minutes on a modest desktop computer [2]. Any dataset of experimental
or simulated crystals can be visualised on maps with continuous and
analytically defined invariant coordinates [3]. These invariants can
be inverted to any generic periodic structure in 3 dimensions,
uniquely under any distance-preserving transformation, and hence
mathematically justify a new rigorous data standard for upholding
scientific integrity in crystallography.
[1] O.Anosova, V.Kurlin, M.Senechal. The importance of definitions in
crystallography. IUCrJ 11 (4), 453-463 (2024).
[2] D.Widdowson, V.Kurlin. Resolving the data ambiguity for periodic
crystals. Advances in Neural Information Processing Systems (NeurIPS),
v.35, p.24625-24638 (2022).
[3] D.Widdowson, V.Kurlin. Continuous invariant-based maps of the
Cambridge Structural Database. Crystal Growth & Design, 24(13),
5627–5636 (2024).

--

T +61 (02) 9717 9907
F +61 (02) 9717 3145
M +61 (04) 0249 4148

_______________________________________________
comcifs mailing list
comcifs@mailman.iucr.org
https://mailman.iucr.org/cgi-bin/mailman/listinfo/comcifs