Tracking the time evolution of soft matter systems via topological structural heterogeneity

Solis, Ingrid Membrillo; Orlova, Tetiana; Bednarska, Karolina; Lesiak, Piotr; Woliński, Tomasz R.; D’Alessandro, Giampaolo; Brodzki, Jacek; Kaczmarek, Malgosia

doi:10.1038/s43246-021-00223-1

Cited by 28 publications

(8 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Smith et al recently used the EC to characterize simulated random fields that contain features that are similar to those of chemoresponsive nematic LC films . Solis et al used persistent homology and a derived structural heterogeneity to track the structural changes in an LC nanocomposite and revealed the effect of confined geometry on the nematic–isotropic and isotropic–nematic phase transition …”

Section: Introductionmentioning

confidence: 99%

Scalable Extraction of Information from Spatiotemporal Patterns of Chemoresponsive Liquid Crystals Using Topological Descriptors

et al. 2023

View full text Add to dashboard Cite

Chemoresponsive liquid crystals (LCs) can be engineered to generate information-rich optical responses (in the form of space–time color and brightness patterns) when they are exposed to target gas contaminants. We investigate the use of topological descriptors (Euler characteristic, lacunarity, and fractal dimension) for extracting different types of information from these complex optical responses and show that these tools can enable the design of sensors and help gain insights into the physical phenomena governing sensor responses. We provide a holistic perspective of topological descriptors using the theory of Minkowski functionals and fractal analysis, which allows us to understand specific information that each descriptor extracts. We also show how to use the topological descriptors in conjunction with space–time filtration operations and color representations to enrich the information extracted. We demonstrate how these capabilities can be used in flexible ways to facilitate unsupervised machine learning (ML) tasks (clustering and visualization) and supervised tasks (regression and classification). We demonstrate the developments using real, high-throughput experimental data sets for functionalized LC films that are exposed to different gaseous environments. We show that the topological descriptors encode significant information and can be used to detect outliers in high-throughput data and visualize the temporal evolution of structure. Moreover, we show that the topological descriptors can be used to predict contaminant concentrations using simple ML models such as support vector machines; notably, these ML models can achieve accuracies comparable to those of powerful convolutional neural networks but with a much lower computational cost (from hours to seconds) and using less sophisticated computing hardware (CPUs instead of GPUs). This scalability enables the analysis of space–time data at high resolutions.

show abstract

Section: Introductionmentioning

confidence: 99%

Scalable Extraction of Information from Spatiotemporal Patterns of Chemoresponsive Liquid Crystals Using Topological Descriptors

et al. 2023

View full text Add to dashboard Cite

show abstract

“…We utilize a compact barcode representation of this information, though alternative representations could include persistence diagrams and persistence landscapes . Analysis of the distribution of topological features can be used to create metrics or descriptors of the imageas has been done in image of liquid crystal nanocomposites or immune cells . Additionally, the vectorized barcode representation is easily implemented in machine learning frameworks for the identification of specific image features …”

Section: Introductionmentioning

confidence: 99%

“…In this work, we consider the surface image obtained by projecting atoms or molecules on the instantaneous surface onto a xy plane (a 2-dim manifold embedded in 3-dim space). Within the topological data analysis (TDA) of manifolds, sublevelset persistent homology (PH) is a valuable tool and has been employed in many studies to characterize image data. − It is particularly adept at identifying minor differences or anomalies within identical images. − Most image analyses have focused on time-independent data; however, in computational chemistry temporal PH descriptors could dramatically enhance our understanding of adsorption/desorption phenomena at surfaces and reactivity.…”

Section: Introductionmentioning

confidence: 99%

Adsorbate Organization Characterized by Sublevelset Persistent Homology

Kumar

Clark

2023

J. Chem. Theory Comput.

View full text Add to dashboard Cite

Interfacial adsorbate organization influences a variety physicochemical properties and reactivity. Surfaces that are rough, defect laden, or have large fluctuations (as in soft matter interfaces) can lead to complex adsorbate structures. This is amplified if adsorbate–adsorbate interactions lead to self-assembly. Although image analysis algorithms are somewhat common for the study of solid interfaces (from microscopy for example), images are often not readily available for adsorbates at soft matter surfaces, and the complexity of adsorbate organization necessitates the development of new characterization approaches. Here we propose the use of adsorbate “density” images from molecular dynamics simulations of liquid/vapor and liquid/liquid interfaces. Topological data analysis is employed to characterize surface active amphiphile self-assembly under nonreactive and reactive conditions. We develop a chemical interpretation of sublevelset persistent homology barcode representations of the density images, in addition to descriptors that clearly differentiate between different reactive and nonreactive organizational regimes. The complexity of amphiphile self-assembly at highly dynamic liquid/liquid interfaces represents a worst-case scenario for adsorbate characterization, and as such the methodology developed is completely generalizable to a wide variety of surface image data, whether from experiment or computer simulation.

show abstract

“…For example, Smith et al recently used the EC to characterize simulated random fields that contain features that are similar to those of chemoresponsive nematic LC films 14 . Solis et al used persistent homology and a derived structural heterogeneity to track the structural changes in an LC nanocomposite and revealed the effect of confined geometry on the nematic-isotropic and isotropic-nematic phase transition 31 .…”

Section: Introductionmentioning

confidence: 99%

Scalable Extraction of Information from Spatio-Temporal Patterns of Chemoresponsive Liquid Crystals Using Topological Descriptors

Jiang

Bao

Smith

et al. 2023

Preprint

View full text Add to dashboard Cite

Chemoresponsive liquid crystals (LCs) can be engineered to generate information-rich optical responses (in the form of space-time color and brightness patterns) when exposed to target gas contaminants. We investigate the use of topological descriptors (Euler characteristic, lacunarity, and fractal dimension) for extracting different types of information from these complex optical responses and show that these tools can enable the design of sensors and help gain insights into physical phenomena governing sensor responses. We provide a holistic perspective of topological descriptors using theory of Minkowski functionals and fractal analysis, which allows us to understand specific information that each descriptor extracts. We also show how to use the topological descriptors in conjunction with space-time filtration operations and color representations to enrich the information extracted. We demonstrate how these capabilities can be used in flexible ways to facilitate unsupervised machine learning (ML) tasks (clustering and visualization) and supervised tasks (regression and classification). We demonstrate the developments using real, high-throughput experimental datasets for functionalized LC films that are exposed to different gaseous environments. We show that the topological descriptors encode significant information and can be used to detect outliers in high-throughput data and visualize the temporal evolution of topology. Moreover, we show that the topological descriptors can be used to predict contaminant concentrations using simple ML models such as support vector machines; notably, these ML models can achieve comparable accuracies to those of powerful convolutional neural networks but with a much lower computational cost (from hours to seconds) and using less sophisticated computing hardware (CPUs instead of GPUs). This scalability enables the analysis of space-time data at high resolutions.

show abstract

Tracking the time evolution of soft matter systems via topological structural heterogeneity

Cited by 28 publications

References 47 publications

Scalable Extraction of Information from Spatiotemporal Patterns of Chemoresponsive Liquid Crystals Using Topological Descriptors

Scalable Extraction of Information from Spatiotemporal Patterns of Chemoresponsive Liquid Crystals Using Topological Descriptors

Adsorbate Organization Characterized by Sublevelset Persistent Homology

Scalable Extraction of Information from Spatio-Temporal Patterns of Chemoresponsive Liquid Crystals Using Topological Descriptors

Contact Info

Product

Resources

About