Computational Challenges Research Articles

Spin Dynamics of Radical Pairs Using the Stochastic Schrödinger Equation in MolSpin.

The chemical reactivity of radical pairs is strongly influenced by the interactions of electronic and nuclear spins. A detailed understanding of these effects requires a quantum description of the spin dynamics that considers spin-dependent reaction rates, interactions with external magnetic fields, spin-spin interactions, and the loss of spin coherence caused by coupling to a fluctuating environment. Modeling real chemical and biochemical systems, which frequently involve radicals with multinuclear spin systems, poses a severe computational challenge. Here, we implement a method based on the stochastic Schrödinger equation in the software package MolSpin. Large electron-nuclear spin systems can be simulated efficiently, with asymmetric spin-selective recombination reactions, anisotropic hyperfine interactions, and nonzero exchange and dipolar couplings. Spin-relaxation can be modeled using the stochastic time-dependence of spin interactions determined by molecular dynamics and quantum chemical calculations or by allowing rate coefficients to be explicitly time-dependent. The flexibility afforded by this approach opens new avenues for exploring the effects of complex molecular motions on the spin dynamics of chemical transformations.

Verification, validation, and uncertainty quantification (VVUQ) in structural analysis of concrete dams

Over the past three decades, advancements in computational power and numerical methods have significantly enhanced the role of structural analyses in the design and safety assessment of dams. Simulating concrete dam behavior, particularly in interactions with reservoir water and rock foundations, poses formidable computational challenges. Additionally, the need to define uncertainties related to material parameters, loading conditions, and modeling strategy adds complexity to the modeling process, therefore, quantifying sources of uncertainty is crucial for maintaining credibility and confidence in analysis results. This paper provides a synthesis and an overview of existing research and presents a generic framework for evaluating the credibility of advanced structural analysis methods for concrete dams, with a focus on their limitations and associated uncertainties. The methodology includes a comprehensive process for structural analysis, verification, validation, and uncertainty quantification, aiming to facilitate condition assessments of concrete dams.

Efficient Visual-Aware Fashion Recommendation Using Compressed Node Features and Graph-Based Learning

In fashion e-commerce, predicting item compatibility using visual features remains a significant challenge. Current recommendation systems often struggle to incorporate high-dimensional visual data into graph-based learning models effectively. This limitation presents a substantial opportunity to enhance the precision and effectiveness of fashion recommendations. In this paper, we present the Visual-aware Graph Convolutional Network (VAGCN). This novel framework helps improve how visual features can be incorporated into graph-based learning systems for fashion item compatibility predictions. The VAGCN framework employs a deep-stacked autoencoder to convert the input image’s high-dimensional raw CNN visual features into more manageable low-dimensional representations. In addition to improving feature representation, the GCN can also reason more intelligently about predictions, which would not be possible without this compression. The GCN encoder processes nodes in the graph to capture structural and feature correlation. Following the GCN encoder, the refined embeddings are input to a multi-layer perceptron (MLP) to calculate compatibility scores. The approach extends to using neighborhood information only during the testing phase to help with training efficiency and generalizability in practical scenarios, a key characteristic of our model. By leveraging its ability to capture latent visual features and neighborhood-based learning, VAGCN thoroughly investigates item compatibility across various categories. This method significantly improves predictive accuracy, consistently outperforming existing benchmarks. These contributions tackle significant scalability and computational efficiency challenges, showcasing the potential transformation of recommendation systems through enhanced feature representation, paving the way for further innovations in the fashion domain.

Sensitivity Analysis of Fatigue Life for Cracked Carbon-Fiber Structures Based on Surrogate Sampling and Kriging Model under Distribution Parameter Uncertainty

The quality and reliability of wind turbine blades, as core components of wind turbines, are crucial for the operational safety of the entire system. Carbon fiber is the primary material for wind turbine blades. However, during the manufacturing process, manual intervention inevitably introduces minor defects, which can lead to crack propagation under complex working conditions. Due to limited understanding and measurement capabilities of the input variables of structural systems, the distribution parameters of these variables often exhibit uncertainty. Therefore, it is essential to assess the impact of distribution parameter uncertainty on the fatigue performance of carbon-fiber structures with initial cracks and quickly identify the key distribution parameters affecting their reliability through global sensitivity analysis. This paper proposes a sensitivity analysis method based on surrogate sampling and the Kriging model to address the computational challenges and engineering application difficulties in distribution parameter sensitivity analysis. First, fatigue tests were conducted on carbon-fiber structures with initial cracks to study the dispersion of their fatigue life under different initial crack lengths. Next, based on the Hashin fatigue failure criterion, a simulation analysis method for the fatigue cumulative damage life of cracked carbon-fiber structures was proposed. By introducing uncertainty parameters into the simulation model, a training sample set was obtained, and a Kriging model describing the relationship between distribution parameters and fatigue life was established. Finally, an efficient input variable sampling method using the surrogate sampling probability density function was introduced, and a Sobol sensitivity analysis method based on surrogate sampling and the Kriging model was proposed. The results show that this method significantly reduces the computational burden of distribution parameter sensitivity analysis while ensuring computational accuracy.

Structural transitions in liquid semiconductor alloys: A molecular dynamics study with a neural network potential.

Liquid-liquid phase transitions hold a unique and profound significance within condensed matter physics. These transitions, while conceptually intriguing, often pose formidable computational challenges. However, recent advances in neural network (NN) potentials offer a promising avenue to effectively address these challenges. In this paper, we delve into the structural transitions of liquid CdTe, CdS, and their alloy systems using molecular dynamics simulations, harnessing the power of an NN potential named LaspNN. Our investigations encompass both pressure and temperature effects. Through our simulations, we uncover three primary liquid structures around melting points that emerge as pressure increases: tetrahedral, rock salt, and close-packed structures, which greatly resemble those of solid states. In the high-temperature regime, we observe the formation of Te chains and S dimers, providing a deeper understanding of the liquid's atomic arrangements. When examining CdSxTe1-x alloys, our findings indicate that a small substitution of S by Te atoms for S-rich alloys (x > 0.5) exhibits a structural transition much different from CdS, while a large substitution of Te by S atoms for Te-rich alloys (x < 0.5) barely exhibits a structural transition similar to CdTe. We construct a schematic diagram for liquid alloys that considers both temperature and pressure, providing a comprehensive overview of the alloy system's behavior. The local aggregation of Te atoms demonstrates a linear relationship with alloy composition x, whereas that of S atoms exhibits a nonlinear one, shedding light on the composition-dependent structural changes.

MetaExpertPro: A Computational Workflow for Metaproteomics Spectral Library Construction and Data-Independent Acquisition Mass Spectrometry Data Analysis

Analysis of large-scale data-independent acquisition mass spectrometry metaproteomics data remains a computational challenge. Here, we present a computational pipeline called metaExpertPro for metaproteomics data analysis. This pipeline encompasses spectral library generation using data-dependent acquisition MS, protein identification and quantification using data-independent acquisition mass spectrometry, functional and taxonomic annotation, as well as quantitative matrix generation for both microbiota and hosts. By integrating FragPipe and DIA-NN, metaExpertPro offers compatibility with both Orbitrap and timsTOF MS instruments. To evaluate the depth and accuracy of identification and quantification, we conducted extensive assessments using human fecal samples and benchmark tests. Performance tests conducted on human fecal samples indicated that metaExpertPro quantified an average of 45,000 peptides in a 60-min diaPASEF injection. Notably, metaExpertPro outperformed three existing software tools by characterizing a higher number of peptides and proteins. Importantly, metaExpertPro maintained a low factual false discovery rate of approximately 5% for protein groups across four benchmark tests. Applying a filter of five peptides per genus, metaExpertPro achieved relatively high accuracy (F-score=0.67-0.90) in genus diversity and showed a high correlation (rSpearman=0.73-0.82) between the measured and true genus relative abundance in benchmark tests. Additionally, the quantitative results at the protein, taxonomy, and function levels exhibited high reproducibility and consistency across the commonly adopted public human gut microbial protein databases IGC and UHGP. In a metaproteomic analysis of dyslipidemia patients, metaExpertPro revealed characteristic alterations in microbial functions and potential interactions between the microbiota and the host.

An efficient physics-dimension-based reduction method for computing frequency response functions of viscoelastically damped systems

The frequency response functions (FRFs) are of critical interest to dynamic problems, however, they suffer from computational challenges for viscoelastically damped systems. In this paper, a physics-space-based reduction method is proposed for predicting the FRFs of large-scale viscoelastically damped systems involving the standard linear solid model. A physics-dimension subspace is constructed based on original system matrices and viscoelastic parameters, which can be easily generated by using a recursive manner. A projection basis generation algorithm is then developed to generate a standard orthonormal basis within the physics-dimension subspace. With the help of the standard orthonormal basis and the moment-matching-based reduction method, a physics-space-based reduction method is proposed for efficiently predicting the FRFs of large-scale viscoelastically damped systems. Unlike the widely used state-space reduction method, the reduced system of the proposed method can preserve system’s physical structure so that the physical meaning can be captured. Using both theoretical and numerical analyses, the proposed physics-space-based method is more accurate and efficient than the state-space-based reduction method.

Neural networks based surrogate modeling for efficient uncertainty quantification and calibration of MEMS accelerometers

This paper addresses the computational challenges inherent in the stochastic characterization and uncertainty quantification of Micro-Electro-Mechanical Systems (MEMS) capacitive accelerometers. Traditional methods, such as Markov Chain Monte Carlo (MCMC) algorithms, are often constrained by the computational intensity required for high-fidelity (e.g., finite element) simulations. To overcome these limitations, we propose to use supervised learning-based surrogate models, specifically artificial neural networks, to effectively approximate the response of MEMS capacitive accelerometers. Our approach involves training the surrogate models with data derived from initial high-fidelity finite element analyses (FEA), providing rich datasets to be generated in an offline phase. The surrogate models replicate the FEA accuracy in predicting the behavior of the accelerometer under a wide range of fabrication parameters, thereby reducing the online computational cost without compromising accuracy. This enables extensive and efficient stochastic analyses of complex MEMS devices, offering a flexible framework for their characterization. A key application of our framework is demonstrated in estimating the sensitivity of an accelerometer, accounting for unknown mechanical offsets, over-etching, and thickness variations. We employ an MCMC approach to estimate the posterior distribution of the device’s unknown fabrication parameters, informed by its response to transient voltage signals. The integration of surrogate models for mapping fabrication parameters to device responses, and subsequently to sensitivity measures, greatly enhances both backward and forward uncertainty quantification, yielding accurate results while significantly improving the efficiency and effectiveness of the characterization process. This process allows for the reconstruction of device sensitivity using only voltage signals, without the need for direct mechanical acceleration stimuli.

Efficient finite element reliability analysis employing sequentially-updated surrogate model for fragility curve derivation

As the threat of natural disasters to structures intensifies, risk assessment of infrastructure has gained much importance. Fragility curves are essential tools in predicting disaster-related losses and making disaster mitigation decisions. In this paper, we propose a new method to efficiently derive accurate fragility curves for structures with high levels of nonlinearity or complexity, addressing the computational challenges of conventional finite element reliability analysis (FERA). To reduce the computational cost for calculating probability of failure in FERA, the proposed method utilizes the first-order reliability method (FORM). However, even with this approach, the computational cost of deriving the fragility curve may remain high; therefore, a surrogate model is used to further reduce costs. By training the surrogate model using the initial structural damage probabilities for a few hazard intensities, an optimal starting point can be calculated for the subsequent FORM analysis. During the fragility analysis, the surrogate model can be updated sequentially to increase the efficiency of FORM analysis continuously. In particular, the training process of the surrogate model requires no separate or additional finite element analysis because it is constructed using previous FERA results. The accuracy and efficiency of the proposed method are tested using conventional FERA and Monte Carlo simulations through a hypothetical short-column example. In addition, fragility curves are derived through a bridge flood fragility assessment considering the scour and seismic vulnerability assessment of a buried gas pipeline considering soil-structure interactions. The derived fragility curves closely match those derived using the conventional FERA, and the computational costs are reduced by 36.54 % and 52.38 %, respectively, compared with the conventional FERA, confirming its cost-effectiveness.

Interstellar computation their challenges and approaches in interplanetary cloud systems and data transmission

With the advancement of human civilisation beyond the confines of Earth's atmosphere, there has been an increasing need for efficient data processing and transmission in extraterrestrial environments. The article explores innovative strategies for resolving intricate computational challenges encountered in the context of interplanetary distances. The purpose of this study is to investigate the impact of issues such as latency, data corruption, and power consumption on classical computing paradigms in space. We examine the concept of "interplanetary cloud systems", which use distributed computation and data storage to tackle these issues. The article also covers advanced techniques for sending data, including communication systems using quantum entanglement and deep space networks. The article integrates the disciplines of astronomy, computer science, and space exploration to provide a holistic perspective on the issue. It is crucial to emphasise that the implementation of fault-tolerant systems, adaptive algorithms, and robust networking protocols is essential for the success of extrasolar computing programs. The article thoroughly explores the potential and risks that arise when space travel, astronomy, and computer science intersect. We explore the potential future of interplanetary data processing and communication technologies via the examination of fault tolerance, adaptive algorithms, and resilient networking. The importance of resolving computational issues in space increases as humanity extends its presence in the cosmos. The study results have significant ramifications for the development of forthcoming technology enabling interplanetary data transfer, hence facilitating human colonisation and scientific exploration beyond Earth.

Bio-Plausible Multimodal Learning with Emerging Neuromorphic Devices.

Multimodal machine learning, as a prospective advancement in artificial intelligence, endeavors to emulate the brain's multimodal learning abilities with the objective to enhance interactions with humans. However, this approach requires simultaneous processing of diverse types of data, leading to increased model complexity, longer training times, and higher energy consumption. Multimodal neuromorphic devices have the capability to preprocess spatio-temporal information from various physical signals into unified electrical signals with high information density, thereby enabling more biologically plausible multimodal learning with low complexity and high energy-efficiency. Here, this work conducts a comparison between the expression of multimodal machine learning and multimodal neuromorphic computing, followed by an overview of the key characteristics associated with multimodal neuromorphic devices. The bio-plausible operational principles and the multimodal learning abilities of emerging devices are examined, which are classified into heterogeneous and homogeneous multimodal neuromorphic devices. Subsequently, this work provides a detailed description of the multimodal learning capabilities demonstrated by neuromorphic circuits and their respective applications. Finally, this work highlights the limitations and challenges of multimodal neuromorphic computing in order to hopefully provide insight into potential future research directions.

Localization–delocalization transition for light particles in turbulence

Small bubbles in fluids rise to the surface due to Archimede's force. Remarkably, in turbulent flows this process is severely hindered by the presence of vortex filaments, which act as moving potential wells, dynamically trapping light particles and bubbles. Quantifying the statistical weights and roles of vortex filaments in turbulence is, however, still an outstanding experimental and computational challenge due to their small scale, fast chaotic motion, and transient nature. Here we show that, under the influence of a modulated oscillatory forcing, the collective bubble behavior switches from a dynamically localized to a delocalized state. Additionally, we find that by varying the forcing frequency and amplitude, a remarkable resonant phenomenon between light particles and small-scale vortex filaments emerges, likening particle behavior to a forced damped oscillator. We discuss how these externally actuated bubbles can be used as a type of microscopic probe to investigate the space-time statistical properties of the smallest turbulence scales, allowing to quantitatively measure physical characteristics of vortex filaments. We develop a superposition model that is in excellent agreement with the simulation data of the particle dynamics which reveals the fraction of localized/delocalized particles as well as characteristics of the potential landscape induced by vortices in turbulence. Our approach paves the way for innovative ways to accurately measure turbulent properties and to the possibility to control light particles and bubble motions in turbulence with potential applications to oceanography, medical imaging, drug/gene delivery, chemical reactions, wastewater treatment, and industrial mixing.

Assessing between-individual variability in bioenergetics modelling: Opportunities, challenges, and potential applications

Population dynamics is influenced by between-individual variability. Dynamic Energy Budget (DEB) theory is an appealing framework for assessing such a variability, yet DEB parameters have rarely been estimated at the individual level. Bayesian hierarchical models show promise for inferring individual variability in DEB parameters, thought computational challenges have limited their use due to the need to solve differential equations. Timely, Stan has emerged as a general-purpose statistical tool for fitting dynamic models. This paper introduces an analytical strategy using Bayesian parametric inference and hierarchical modelling to estimate individual-specific DEB parameters. Two biologically relevant DEB parameters were successfully estimated for 69 Gilt-head breams (Sparus aurata) with up to 11 measures of length and wet weight each. The estimated between-individual variability in these two DEB parameters explained well the observed patterns in length and weight at between- and within-individual levels. Moreover, data-simulation experiments highlighted the potential and limitations of our approach, suggesting that improved data collection could enable to increase precision and the number of DEB parameters that can be estimated at the individual level. This strategy can better represent between-individual variability in DEB parameters, which ultimately may improve forecasting of population dynamics after integrating DEB into population models.

The use of convolutional neural networks for abnormal behavior recognition in crowd scenes

This study introduces the Abnormality Converging Scene Analysis Method (ACSAM) to detect abnormal group behavior using monitored videos or CCTV images in crowded scenarios. Abnormal behavior recognition involves classifying activities and gestures in continuous scenes, which traditionally presents significant computational challenges, particularly in complex crowd scenes, leading to reduced recognition accuracy. To address these issues, ACSAM employs a convolutional neural network (CNN) enhanced with Abnormality and Crowd Behavior Training layers to accurately detect and classify abnormal activities, regardless of crowd density. The method involves extracting frames from the input scene and using CNN to perform conditional validation of abnormality factors, comparing current values with previous high values to maximize detection accuracy. As the abnormality factor increases, the identification rate improves with higher training iterations. The system was tested on 26 video samples and trained on 34 samples, demonstrating superior performance to other approaches like DeepROD, MSI-CNN, and PT-2DCNN. Specifically, ACSAM achieved a 12.55% improvement in accuracy, a 12.97% increase in recall, and a 10.23% enhancement in convergence rate. These results suggest that ACSAM effectively overcomes the computational challenges inherent in crowd scene detection, offering a robust solution for real-time abnormal behavior recognition in crowded environments.

Extra trees regression assisted 1D monolith reactor simulations based on microkinetic analysis and rate transformation

The incorporation of microkinetic method for surface reaction calculation is a key development direction in monolith reactor simulations, albeit posing significant computational stability and efficiency challenges. A data-driven multi-scale simulation framework is proposed to bridge the gap between macroscopic numerical model and microscopic reaction solvers. After the generating of precomputed microkinetic dataset, we utilize a tailor-designed preprocessing transformation, AD_log10, to effectively manage the common challenges of large variances and reversed directions in microkinetic rate data. A suite of machine learning regression models is developed as reaction rate solver integrated into the 1D two-phase monolith model. We employ NH3 selective catalytic oxidation on Cu(100) and Cu(111) as probe systems to validate our framework’s efficacy. In the context of on-board simulations under operating conditions, the Extra Trees Regression (with best regression performance) based model exhibits commendable accuracy and efficiency, successfully addressing complex processes previously unmanageable with interpolation-based models. This innovative approach paves the way for advanced monolith reactor modeling and sheds light on multi-scale simulation methodologies in chemical reaction engineering.

A Clustering Pruning Method Based on Multidimensional Channel Information

Pruning convolutional neural networks offers a promising solution to mitigate the computational complexity challenges encountered during application deployment. However, prevalent pruning techniques primarily concentrate on model parameters or feature mapping analysis to devise static pruning strategies, often overlooking the underlying feature extraction capacity of convolutional kernels. To address this, the study first quantitatively expresses the feature extraction capability of convolutional channels from three aspects: global features, distribution metrics, and directional metrics. It explores the multi-dimensional information of the channels, calculates the overall expectation, variance, and cosine distance from the unit vector as the quantitative results of the channels. Subsequently, a clustering algorithm is employed to categorize the multidimensional information. This approach ensures that convolutional channels grouped within each cluster possess similar feature extraction capabilities. An enhanced differential evolutionary algorithm is utilized to optimize the number of clustering centers across all convolutional layers, ensuring optimal grouping. The final step involves achieving channel sparsification through the calculation of crowding distances for each sample within its designated cluster. This preserves a diverse subset of channels that are critical for maintaining model accuracy. Extensive empirical evaluations conducted on three benchmark image classification datasets demonstrate the efficacy of this method. For instance, on the ImageNet dataset, the ResNet-50 model experiences a substantial reduction in FLOPs by 58.43% while incurring a minimal decrease in TOP-1 accuracy of only 1.15%.

Pre-disaster allocation and post-disaster dispatch strategies of power emergency resources for resilience enhancement of distribution networks

With the frequent occurrence of large-scale power outages caused by extreme disasters, coordinated dispatch of power emergency resources (PERs) and repair crews (RCs) has gradually become an essential method to enhance the resilience of distribution networks (DNs). Previous works solve this coordinated dispatch problem in a risk-neutral approach with the assumption that the supply of power repair materials (PRMs) is still fully available after the disaster. However, growing evidence indicates that certain extreme disasters may lead to limited availability of PERs in the DN, rendering risk-neutral decision making impractical. To fill this gap, this paper proposes a resilience-oriented PERs pre-disaster allocation and post-disaster dispatch strategy in a risk-averse manner. In the pre-disaster prevention process, a two-stage distributionally robust optimization (DRO) model based on mean-CVaR is developed, considering the uncertainties of both system damage and PERs demands, along with the decision maker's preference for risk. In the post-disaster restoration process, a collaborative spatio-temporal dispatching model for PERs and RCs with the goal of minimizing outage losses is established. To tackle the computational challenges associated with PRMs allocation based on DRO, robust counterpart transformation and dual theory are utilized. Case studies are conducted using a real world modified 118-bus distribution network. In comparative analysis, five control groups are established to validate the necessity to consider the limited supply of PRMs and the risk-averse attitude of decision makers in enhancing DNs resilience.

FINEST: Stabilizing Recommendations by Rank-Preserving Fine-Tuning

Modern recommender systems may output considerably different recommendations due to small perturbations in the training data. Changes in the data from a single user will alter the recommendations as well as the recommendations of other users. In applications like healthcare, housing, and finance, this sensitivity can have adverse effects on user experience. We propose a method to stabilize a given recommender system against such perturbations. This is a challenging task due to (1) the lack of a “reference” rank list that can be used to anchor the outputs; and (2) the computational challenges in ensuring the stability of rank lists with respect to all possible perturbations of training data. Our method, FINEST, overcomes these challenges by obtaining reference rank lists from a given recommendation model and then fine-tuning the model under simulated perturbation scenarios with rank-preserving regularization on sampled items. Our experiments on real-world datasets demonstrate that FINEST can ensure that recommender models output stable recommendations under a wide range of different perturbations without compromising next-item prediction accuracy.

Advancing regional flood mapping in a changing climate: A HAND‐based approach for New Jersey with innovations in catchment analysis

AbstractRegional flood mapping poses computational and spatial heterogeneity challenges, exacerbated by climate change‐induced uncertainties. This study focuses on creating a state‐wide flood mapping solution with enhanced accuracy and computational speed to support regional flooding hazard analysis and the assessment of climate change, using New Jersey as a case study. The Height Above Nearest Drainage (HAND) framework was employed for large‐scale flood mapping. The model was validated against high water marks (HWMs) collected after Hurricane Irene. Based on the National Water Model (NWM), synthetic rating curves in HAND were calibrated by tuning Manning's roughness, aligning the predicted and observed flood depths. The roughness values were generalized across the state from the validated water basins to the ungauged ones, using a multivariate regression with the hydrologic and geographic information. To map the future climate‐change‐induced flooding, a correlation between NOAA historical precipitation totals and NWM flow data from 2010 to 2020 was established to link precipitation and runoff. This study also invented a novel method for correcting catchment discontinuities, inherent in the HAND model, based on a computer vision scheme, the Sobel filter. The modeling results show that average and worst‐case storm events have the potential to increase 10%–50% in the state, where mountain areas and major river banks would be exposed to this impact more significantly.

Enhancing Mass-Cytometry Imaging Analysis

Mass cytometry imaging is a technique that utilizes mass spectrometry instead of fluorescent flow cytometry to detect metal-conjugated antibodies binding to single cell antigens. This method generates multi-dimensional images, a complex data type with significant computational and interpretation challenges due to the absence of a standardized analysis workflow. To address this, our project evaluated the quality and efficacy of a previous study’s proposed workflow to advance the interpretability and accuracy of mass cytometry data analysis. The approach evaluated whether the previously proposed workflow could effectively analyze alternative samples classified by previous studies as “mixed” immune structure. The methods included an in-depth examination of cell labeling based on spatial proximity, which emerged as a secondary analytical outcome. This research advances spatial labeling by enhancing the spatial information related to physical connections between interacting cells and incorporating biological interactions, such as protein involvement. By integrating these factors, this research proposes a method for more accurate labeling, contributing to a more robust analysis workflow for this complex data type. Unstructured “mixed” samples remained distinguishable, though less effectively than structured “compartmentalized” samples. Future applications involve exploring additional datasets to further develop means of distinguishing between contact and non-contact interactions among cancer and immune cells.