A5067536683- Topological and Geometric Data Analysis
- Geometric and Algebraic Topology
- Machine Learning in Materials Science
- X-ray Diffraction in Crystallography
- Computational Geometry and Mesh Generation
- Computational Drug Discovery Methods
- Digital Image Processing Techniques
- Homotopy and Cohomology in Algebraic Topology
- Advanced Image and Video Retrieval Techniques
- Cell Image Analysis Techniques
- Quasicrystal Structures and Properties
- Advanced Combinatorial Mathematics
- Data Management and Algorithms
- Crystallography and molecular interactions
- Crystallization and Solubility Studies
- Advanced Topics in Algebra
- Geochemistry and Geologic Mapping
- Algorithms and Data Compression
- Algebraic structures and combinatorial models
- Advanced Graph Theory Research
- Medical Image Segmentation Techniques
- History and advancements in chemistry
- Meteorological Phenomena and Simulations
- Protein Structure and Dynamics
- Data Visualization and Analytics
University of Liverpool
2016-2025
Microsoft Research (United Kingdom)
2015-2016
Durham University
2006-2015
Microsoft (United States)
2015
Institut de Mathématiques de Bourgogne
2004-2005
Université de Bourgogne
2004-2005
Institut de Mathématiques de Bordeaux
2005
Independent University of Moscow
2005
Lomonosov Moscow State University
1999-2004
Abstract. The Atmospheric River Tracking Method Intercomparison Project (ARTMIP) is an international collaborative effort to understand and quantify the uncertainties in atmospheric river (AR) science based on detection algorithm alone. Currently, there are many AR identification tracking algorithms literature with a wide range of techniques conclusions. ARTMIP strives provide community information different methodologies guidance most appropriate for given question or region interest. All...
Abstract Atmospheric rivers (ARs) are now widely known for their association with high‐impact weather events and long‐term water supply in many regions. Researchers within the scientific community have developed numerous methods to identify track of ARs—a necessary step analyses on gridded data sets, objective attribution impacts ARs. These different been answer specific research questions hence use criteria (e.g., geometry, threshold values key variables, time dependence). Furthermore,...
Mesoporous molecular crystals have potential applications in separation and catalysis, but they are rare hard to design because many weak interactions compete during crystallization, most molecules an energetic preference for close packing. Here, we combine crystal structure prediction (CSP) with structural invariants continuously qualify the similarity between predicted structures related molecules. This allows isomorphous substitution strategies, which can be unreliable crystals, augmented by
Abstract The application of machine learning models to predict material properties is determined by the availability high-quality data. We present an expert-curated dataset lithium ion conductors and associated conductivities measured a.c. impedance spectroscopy. This has 820 entries collected from 214 sources; contain a chemical composition, expert-assigned structural label, ionic conductivity at specific temperature (from 5 873 °C). There are 403 unique compositions with near room (15–35...
Abstract The structure–property hypothesis says that the properties of all materials are determined by an underlying crystal structure. main obstacle was ambiguity conventional representations based on incomplete or discontinuous descriptors allow false negatives positives. This resolved ultra-fast pointwise distance distribution, which distinguished periodic structures in world’s largest collection real (Cambridge structural database). State-of-the-art results property prediction were...
It is a core problem in any field to reliably tell how close two objects are being the same, and once this relation has been established, we can use information precisely quantify potential relationships, both analytically with machine learning (ML). For inorganic solids, chemical composition fundamental descriptor, which be represented by assigning ratio of each element material vector. These vectors convenient mathematical data structure for measuring similarity, but unfortunately,...
Abstract. Identifying weather patterns that frequently lead to extreme events is a crucial first step in understanding how they may vary under different climate change scenarios. Here, we propose an automated method for recognizing atmospheric rivers (ARs) data using topological analysis and machine learning. The provides useful information about features (shape characteristics) statistics of ARs. We illustrate this by applying it outputs version 5.1 the Community Atmosphere Model (CAM5.1)...
The fundamental model of any solid crystalline material (crystal) at the atomic scale is a periodic point set. strongest natural equivalence crystals rigid motion or isometry that preserves all inter-atomic distances. Past comparisons structures often used manual thresholds, symmetry groups and reduced cells, which are discontinuous under perturbations thermal vibrations atoms. This work defines infinite sequence continuous invariants (Average Minimum Distances) to progressively capture...
Periodic material or crystal property prediction using machine learning has grown popular in recent years as it provides a computationally efficient replacement for classical simulation methods. A crucial first step any of these algorithms is the representation used periodic crystal. While similar objects like molecules and proteins have finite number atoms their can be built based upon point cloud interpretation, crystals are unbounded size, making more challenging. In present work, we...
Proteins are large biomolecules that regulate all living organisms and consist of one or several chains. The primary structure a protein chain is sequence amino acid residues whose three main atoms (alpha-carbon, nitrogen, carbonyl carbon) form backbone. tertiary the rigid shape represented by atomic positions in 3-dimensional space. Because different geometric structures often have distinct functional properties, it important to continuously quantify differences shapes backbones....
A global analysis of protein crystal structures in the Protein Data Bank (PDB) using a newly developed computational approach reveals many pairs with (nearly) identical main-chain coordinates. Such cases are identified and analyzed, showing that duplication is possible since PDB does not currently have tools or mechanisms would detect potentially duplicate submissions. Some duplicated entries represent modeling efforts ligand binding masquerade as experimentally determined structures. We...
Rigid structures such as cars or any other solid objects are often represented by finite clouds of unlabeled points. The most natural equivalence on these point is rigid motion isometry maintaining all inter-point distances. patterns can be reliably compared only complete invariants that also called equivariant descriptors without false negatives (isometric having different descriptions) and positives (non-isometric with the same description). Noise in data motivate a search for continuous...
Abstract Periodic material or crystal property prediction using machine learning has grown popular in recent years as it provides a computationally efficient replacement for classical simulation methods. A crucial first step any of these algorithms is the representation used periodic crystal. While similar objects like molecules and proteins have finite number atoms their can be built based upon point cloud interpretation, crystals are unbounded size, making more challenging. In present...
Abstract Real data are often given as a noisy unstructured point cloud, which is hard to visualize. The important problem represent topological structures hidden in cloud by using skeletons with cycles. All past skeletonization methods require extra parameters such scale or noise bound. We define homologically persistent skeleton, depends only on of points and contains optimal subgraphs representing 1‐dimensional cycles the across all scales. full skeleton universal structure encoding...
The packaging industry faces mounting demand to integrate post-consumer recyclate (PCR). However, the complex structure-property relationships of PCRs often obscure their performance compared virgin equivalents, posing challenges in selecting suitable for applications. Focused on extrusion blow moulding grade high-density polyethylene (HDPE), this study presents most extensive characterisation HDPE PCR date, encompassing 23 resins (3 virgin, 20 PCR). Employing Fourier-transform infrared...
Abstract Persistent homology is a popular and useful tool for analysing finite metric spaces, revealing features that can be used to distinguish sets of unlabeled points as input into machine learning pipelines. The famous stability theorem persistent provides an upper bound the change persistence in bottleneck distance under perturbations points, but without giving lower bound. This paper clarifies possible limitations may have distinguishing which evident non-isometric point with identical...
Molecular set transformer is a deep learning architecture for scoring molecular pairs found in co-crystals, whilst tackling the class imbalance problem observed on datasets that include only successful synthetic attempts.
Abstract A periodic lattice in Euclidean space is the infinite set of all integer linear combinations basis vectors. Any can be generated by infinitely many different bases. This ambiguity was partially resolved, but standard reductions remain discontinuous under perturbations modelling atomic displacements. paper completes a continuous classification 2-dimensional lattices up to isometry (or congruence), rigid motion (without reflections), and similarity (with uniform scaling). The new...
This paper was motivated by the articles `Same or different – that is question' in CrystEngComm (July 2020) and `Change to definition of a crystal' IUCr Newsletter (June 2021). Experimental approaches crystal comparisons require rigorously defined classifications crystallography beyond. Since structures are determined rigid form, their strongest equivalence practice motion, which composition translations rotations 3D space. Conventional representations based on reduced cells standardizations...
The hexabasic book is the cone of 1-dimensional skeleton union two tetrahedra glued along a common face.The universal 3-dimensional polyhedron UP product segment and book.We show that any closed 2-dimensional surface in 4-space isotopic to UP.The proof based on representation surfaces by marked graphs, links with double intersections 3-space.We construct finitely presented semigroup whose central elements uniquely encode all isotopy classes surfaces.
Abstract This paper develops a new continuous approach to similarity between periodic lattices of ideal crystals. Quantifying crystal structures is needed substantially speed up the structure prediction, because prediction many target properties computationally slow and essentially repeated for nearly identical simulated structures. The proposed distances arbitrary are invariant under all rigid motions, satisfy metric axioms continuity atomic perturbations. above make these tools clustering...
Abstract Periodic Geometry studies isometry invariants of periodic point sets that are also continuous under perturbations. The motivations come from crystals whose structures determined in a rigid form, but any minimal cells can discontinuously change due to small noise measurements. For integer $$k\ge 0$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mi>k</mml:mi> <mml:mo>≥</mml:mo> <mml:mn>0</mml:mn> </mml:mrow> </mml:math> , the density function set S was...
The most fundamental model of a molecule is cloud unordered atoms, even without chemical bonds that can depend on thresholds for distances and angles. strongest equivalence between clouds atoms rigid motion, which composition translations rotations. existing datasets experimental simulated molecules require continuous quantification similarity in terms distance metric. While m ordered points were continuously classified by Lagrange’s quadratic forms (distance matrices or Gram matrices),...