Search

Summary Imagine if we could simultaneously predict spatial protein expression in tissues from their routine Hematoxylin and Eosin (H&E) stained images, and create tissue images given protein expression profiles thus enabling virtual simulations of how protein expression alterations impact histology in complex diseases like cancer. Such an approach could lead to more informed diagnostic and therapeutic decisions for precision medicine at lower costs and shorter turnaround times, more detailed insights into underlying disease pathology as well as improvement in predictive and generative performance. In this study, we investigate the intricate correlation between protein expressions obtained from Hyperion mass cytometry and histopathological microstructures in conventional H&E stained glioblastoma (GBM) samples, unveiling morphological patterns and cellular-level spatial alterations associated with protein expression changes. To model these complex relationships, we propose a novel generative-predictive framework called Ouroboros for producing H&E images from protein expressions and simultaneously predicting protein expressions from H&E images. Our comprehensive sample-independent validation over 9920 tissue spots from 4 GBM samples encompassing visual image analysis, quantitative analysis, subspace alignment and perturbation experiments shows that the proposed generative-predictive approach offers significant improvements in predicting protein expression from images in comparison to baseline methods as well as accurate generation of virtual GBM sample images. This proof of concept study can contribute to advancing our understanding of histological responses to protein expression perturbations and lays the foundations for further developments in this area. Availability and implementation Implementation and associated data for the proposed approach are available at the URL: https://github.com/Srijay/Ouroboros. [ABSTRACT FROM AUTHOR]

Accretion discs in active galactic nuclei (AGNs) foster black hole (BH) formation, growth, and mergers. Stellar mass BHs migrate inwards under the influence of hydrodynamical torques unless they encounter a region where the torque flips sign. At these migration traps, BHs accumulate and merge via dynamical or gas-assisted interactions, producing high-frequency LIGO/Virgo/KAGRA (LVK) gravitational wave (GW) sources and potentially cutting off the supply of extreme mass ratio inspirals that would otherwise make low-frequency, LISA -band GWs. In this paper, we study the interplay between different types of migration torques, focusing especially on the 'thermal torques' generated by the thermal response of the AGN to embedded stellar-mass BHs that accrete through their own mini-discs. In contrast to previous work, we find that Type I torques cannot produce migration traps on their own, but thermal torques often do, particularly in low-mass AGN. The migration traps produced by thermal torques exist at much larger distances (∼103−5 gravitational radii) than do previously identified Type I traps, carrying implications for GW populations at multiple frequencies. Finally, we identify a bifurcation of AGN discs into two regimes: migration traps exist below a critical AGN luminosity, and do not at higher luminosities. This critical luminosity is fit as |$\log _{10} L_{\rm AGN}^c = 45 {\!-\!} 0.32 \log _{10}{(\alpha /0.01)}$| where α is the Shakura–Sunyaev viscosity parameter, a range compatible with recent claims that LVK GWs are not preferentially associated with high-luminosity AGN. [ABSTRACT FROM AUTHOR]

In this paper, we continue our analysis of the chaotic four-body problem and our study of binary–binary interactions in star clusters. We present a general ansatz-based analytical treatment using statistical mechanics, where each outcome of the four-body problem is regarded as some variation of the three-body problem. For example, when two single stars are produced (the 2 + 1 + 1 outcome), each ejection event is modelled as its own three-body interaction by assuming that the ejections are well separated in time. This is a generalization of the approach adopted in Paper I, based on the density-of-states formalism. There are three possible outcomes for the four-body problem with negative total energies: 2 + 2, 2 + 1 + 1, and 3 + 1. For each outcome, we apply an ansatz-based approach to deriving analytical distribution functions that describe the properties of the products of chaotic four-body interactions involving point particles. To test our theoretical distributions, we perform a set of scattering simulations in the equal-mass point-particle limit using FEWBODY , where we vary the initial ratio of binary semimajor axes. We compare our final theoretical distributions to the simulations for each particular scenario, finding consistently good agreement between the two. The highlights of our results include that binary–binary scatterings act to systematically destroy binaries producing instead a binary and two ejected stars (when the initial binary semimajor axes are similar) or a stable triple (when the initial semimajor axes are very different). The 2 + 2 outcome produces the widest binaries, and the 2 + 1 + 1 outcome produces the most compact binaries. [ABSTRACT FROM AUTHOR]

A star wandering close enough to a massive black hole can be ripped apart by the tidal forces of the black hole. The advent of wide-field surveys at many wavelengths has quickly increased the number of tidal disruption events (TDEs) observed, and has revealed that (i) observed TDE rates are lower than theoretical predictions and (ii) E+A galaxies are significantly overrepresented. This overrepresentation further worsens the tension between observed and theoretically predicted TDEs for non-E+A galaxies. Classical loss cone theory focuses on the cumulative effect of many weak scatterings. However, a strong scattering can remove a star from the distribution before it can get tidally disrupted. Most stars undergoing TDEs come from within the radius of influence, the densest environments of the Universe. In such environments, close encounters rare elsewhere become non-negligible. We revise the standard loss cone theory to take into account classical two-body interactions as well as strong scattering, collisions, tidal captures, and study under which conditions close encounters can shield the loss cone. We (i) analytically derive the impact of strong scattering and other close encounters, (ii) compute time-dependent loss cone dynamics including both weak and strong encounters, and (iii) derive analytical solutions to the Fokker–Planck equation with strong scattering. We find that (i) TDE rates can be reduced to up to an order of magnitude and (ii) strong shielding preferentially reduces deeply plunging stars. We also show that stellar overdensities, one possible explanation for the E + A preference, can fail to increase TDE rates when taking into account strong scattering. [ABSTRACT FROM AUTHOR]

There is increasing interest in the application of Raman spectroscopy in a medical setting, ranging from supporting real-time clinical decisions e.g. surgical margins to assisting pathologists with disease classification. However, there remain a number of barriers for adoption in the medical setting due to the increased complexity of probing highly heterogeneous, dynamic biological materials. This inherent challenge can also limit the deployment of higher level analytical approaches such as Artificial Intelligence (AI) including convolutional neural networks (CNN), as there is a lack of a ground truth required for training purposes i.e. in complex clinical samples. Principal component analysis (PCA) is an unsupervised data reduction approach (orthogonal linear transformation) that has been used extensively in spectroscopy for 30+ years, due to its capability to simplify analysis of complex spectroscopic data. However, due to PCA being unsupervised features will inherently appear mixed and their rank may vary between experiments. Here we propose Guided PCA (GPCA), a simple approach that allows PCA to be guided with spectral data to ensure a consistent rank of a key target moiety by the inclusion of a reference (guiding) spectrum to the data set. This simplifies analysis, increases robustness of PCA analysis and improves quantification and the limits of detection and decreases RMSE. [ABSTRACT FROM AUTHOR]

Noninvasive detection of surface-enhanced Raman spectroscopy (SERS) signals from deep within tissue represents a common challenge in many biological and clinical applications including disease diagnosis and therapy monitoring. Such signals are typically weak and not readily discernible from often much larger Raman and fluorescence background signals (e.g., from surrounding tissue). Consequently, suboptimal sensitivity in the detection of SERS signals is often achieved in these situations. Similar issues can arise in SERS measurements in other diffusely scattering samples and complex matrices. Here, we propose a novel concept, active SERS, for the efficient retrieval of SERS signals from deep within complex matrices such as biological tissues that mitigates these issues. It relies on applying an external perturbation to the sample to alter the SERS signal from nanoparticles (NPs) deep inside the matrix. A measurement with and without, or before and after, such perturbation then can provide powerful contrasting data enabling an effective elimination of the matrix signals to reveal more clearly the desired SERS signal without the interfering background and associated artifacts. The concept is demonstrated using ultrasound (US) as an external source of perturbation and SERS NPs inserted deep within a heterogeneous tissue phantom mimicking a cluster of NPs accumulated within a small target lesion. The overall SERS signal intensity induced by the applied US perturbation decreased by ∼21% and the SERS signal contrast was considerably improved by eliminating subtraction artifacts present in a conventional measurement performed at a neighboring spatial location in a heterogeneous tissue sample. Although the technique was demonstrated with SERS gold NPs with a standard Raman label, it is envisaged that active SERS NPs (both the nanoscale metal geometry and Raman label) could be specifically designed to deliver an augmented response to the external stimulus to further enhance the achievable SERS signal contrast and yield even greater improvement in detection sensitivity. The method was demonstrated using transmission Raman spectroscopy; however, it is also applicable to other Raman implementations including spatially offset Raman spectroscopy and conventional Raman spectroscopy performed both at depth and at surfaces of complex matrices. [ABSTRACT FROM AUTHOR]