Search

Images are increasingly becoming the currency for documenting biodiversity on the planet, providing novel opportunities for accelerating scientific discoveries in the field of organismal biology, especially with the advent of large vision-language models (VLMs). We ask if pre-trained VLMs can aid scientists in answering a range of biologically relevant questions without any additional fine-tuning. In this paper, we evaluate the effectiveness of 12 state-of-the-art (SOTA) VLMs in the field of organismal biology using a novel dataset, VLM4Bio, consisting of 469K question-answer pairs involving 30K images from three groups of organisms: fishes, birds, and butterflies, covering five biologically relevant tasks. We also explore the effects of applying prompting techniques and tests for reasoning hallucination on the performance of VLMs, shedding new light on the capabilities of current SOTA VLMs in answering biologically relevant questions using images. The code and datasets for running all the analyses reported in this paper can be found at https://github.com/sammarfy/VLM4Bio., Comment: 36 pages, 37 figures, 7 tables

This study presents a novel method for the calculation of void fractions in the Computational Fluid Dynamics-Discrete Element Method (CFD-DEM) simulations known as the Gaussian Integral Method (GIM). GIM demonstrates versatility across different types of grids including structured and unstructured polyhedral meshes without the need for any treatment at the boundaries. An optimization method is developed for making GIM grid-independent from resolution and type. The method is validated against experimental data of a fluidized bed. The validation results demonstrate realistic results which resemble experimental data. Furthermore, unstructured polyhedral grids outperformed structured grids of equivalent resolution providing results more aligned with experimental data. The gradient of the void fraction is solved in the CFD solver and is used in the DEM solver for precise estimation of the void fraction at a particle location. In conclusion, GIM provides a solution for void fraction calculations in particulate media simulations with complex geometries applicable to industrial needs.

A central problem in biology is to understand how organisms evolve and adapt to their environment by acquiring variations in the observable characteristics or traits of species across the tree of life. With the growing availability of large-scale image repositories in biology and recent advances in generative modeling, there is an opportunity to accelerate the discovery of evolutionary traits automatically from images. Toward this goal, we introduce Phylo-Diffusion, a novel framework for conditioning diffusion models with phylogenetic knowledge represented in the form of HIERarchical Embeddings (HIER-Embeds). We also propose two new experiments for perturbing the embedding space of Phylo-Diffusion: trait masking and trait swapping, inspired by counterpart experiments of gene knockout and gene editing/swapping. Our work represents a novel methodological advance in generative modeling to structure the embedding space of diffusion models using tree-based knowledge. Our work also opens a new chapter of research in evolutionary biology by using generative models to visualize evolutionary changes directly from images. We empirically demonstrate the usefulness of Phylo-Diffusion in capturing meaningful trait variations for fishes and birds, revealing novel insights about the biological mechanisms of their evolution.

Fishes are integral to both ecological systems and economic sectors, and studying fish traits is crucial for understanding biodiversity patterns and macro-evolution trends. To enable the analysis of visual traits from fish images, we introduce the Fish-Visual Trait Analysis (Fish-Vista) dataset - a large, annotated collection of about 60K fish images spanning 1900 different species, supporting several challenging and biologically relevant tasks including species classification, trait identification, and trait segmentation. These images have been curated through a sophisticated data processing pipeline applied to a cumulative set of images obtained from various museum collections. Fish-Vista provides fine-grained labels of various visual traits present in each image. It also offers pixel-level annotations of 9 different traits for 2427 fish images, facilitating additional trait segmentation and localization tasks. The ultimate goal of Fish-Vista is to provide a clean, carefully curated, high-resolution dataset that can serve as a foundation for accelerating biological discoveries using advances in AI. Finally, we provide a comprehensive analysis of state-of-the-art deep learning techniques on Fish-Vista.

Arctic sea ice plays integral roles in both polar and global environmental systems, notably ecosystems, communities, and economies. As sea ice continues to decline due to climate change, it has become imperative to accurately predict the future of sea ice extent (SIE). Using datasets of Arctic meteorological and SIE variables spanning 1979 to 2021, we propose architectures capable of processing multivariate time series and spatiotemporal data. Our proposed framework consists of ensembled stacked Fourier Transform signals (FFTstack) and Gradient Boosting models. In FFTstack, grid search iteratively detects the optimal combination of representative FFT signals, a process that improves upon current FFT implementations and deseasonalizers. An optimized Gradient Boosting Regressor is then trained on the residual of the FFTstack output. Through experiment, we found that the models trained on both multivariate and spatiotemporal time series data performed either similar to or better than models in existing research. In addition, we found that integration of FFTstack improves the performance of current multivariate time series models. We conclude that the high flexibility and performance of this methodology have promising applications in guiding future adaptation, resilience, and mitigation efforts in response to Arctic sea ice retreat., Comment: Published at IEEE ICMLA 2023 REU Symposium. This version includes minor updates made in the text after original publication

Noncentrosymmetric superconductors can support flat bands of zero-energy surface states in part of their surface Brillouin zone. This requires that they obey time-reversal symmetry and have a sufficiently strong triplet-to-singlet-pairing ratio to exhibit nodal lines in the bulk. These bands are protected by a winding number that relies on chiral symmetry, which is realized as the product of time-reversal and particle-hole symmetry. We reveal a way to stabilize a flat band in the entire surface Brillouin zone, while the bulk dispersion is fully gapped. This idea could lead to a robust platform for quantum computation and represents an alternative route to strongly correlated flat bands in two dimensions, besides twisted bilayer graphene. The necessary ingredient is an additional spin-rotation symmetry that forces the direction of the spin-orbit-coupling vector not to depend on the momentum component normal to the surface. We define a winding number that leads to flat zero-energy surface bands due to bulk-boundary correspondence. We discuss under which conditions this winding number is nonzero in the entire surface Brillouin zone and verify the occurrence of zero-energy surface states by exact numerical diagonalization of the Bogoliubov-de Gennes Hamiltonian for a slab. In addition, we consider how a weak breaking of the additional symmetry affects the surface band, employing first-order perturbation theory and a quasiclassical approximation. We find that the surface states and the bulk gap persist for weak breaking of the additional symmetry but that the band does not remain perfectly flat. The broadening of the band strongly depends on the deviation of the spin-orbit-coupling vector from its unperturbed direction as well as on the spin-orbit-coupling strength and the triplet-pairing amplitude., Comment: 18 pages, 6 figures

Discovering evolutionary traits that are heritable across species on the tree of life (also referred to as a phylogenetic tree) is of great interest to biologists to understand how organisms diversify and evolve. However, the measurement of traits is often a subjective and labor-intensive process, making trait discovery a highly label-scarce problem. We present a novel approach for discovering evolutionary traits directly from images without relying on trait labels. Our proposed approach, Phylo-NN, encodes the image of an organism into a sequence of quantized feature vectors -- or codes -- where different segments of the sequence capture evolutionary signals at varying ancestry levels in the phylogeny. We demonstrate the effectiveness of our approach in producing biologically meaningful results in a number of downstream tasks including species image generation and species-to-species image translation, using fish species as a target example.

In this paper, we introduce PathRTM, a novel deep neural network detector based on RTMDet, for automated KI-67 proliferation and tumor-infiltrated lymphocyte estimation. KI-67 proliferation and tumor-infiltrated lymphocyte estimation play a crucial role in cancer diagnosis and treatment. PathRTM is an extension of the PathoNet work, which uses single pixel keypoints for within each cell. We demonstrate that PathRTM, with higher-level supervision in the form of bounding box labels generated automatically from the keypoints using NuClick, can significantly improve KI-67 proliferation and tumorinfiltrated lymphocyte estimation. Experiments on our custom dataset show that PathRTM achieves state-of-the-art performance in KI-67 immunopositive, immunonegative, and lymphocyte detection, with an average precision (AP) of 41.3%. Our results suggest that PathRTM is a promising approach for accurate KI-67 proliferation and tumor-infiltrated lymphocyte estimation, offering annotation efficiency, accurate predictive capabilities, and improved runtime. The method also enables estimation of cell sizes of interest, which was previously unavailable, through the bounding box predictions., Comment: 12 pages, 11 figures

Pennsylvania has "the highest cyber charter school enrollment in the country" with student enrollment soaring in 2020-21 due to concerns related to the COVID-19 pandemic. School districts paid over $1 billion in tuition for students enrolled in Pennsylvania's 14 cyber charter schools in 2020-21, a $335 million increase over the prior school year. Districts now cite charter school tuition as their top budget pressure. In a 2017 study, Research for Action found that districts are unable to realize substantial savings when students depart for charter schools, leaving the vast majority of tuition as a "stranded cost" or negative fiscal impact on district finances. Here we apply the findings of that study to estimate the fiscal impact of the expansion of cyber charter schools attributable to the pandemic. We find that year-one stranded costs ranged from $290 to $308 million dollars statewide, highlighting the need for policymakers to address the inefficient spending on cyber charter schools in current law. We also describe the broader context of cyber charter schools in Pennsylvania.

Surfaces of nodal noncentrosymmetric superconductors can host flat bands of Majorana modes, which provide a promising platform for quantum computation if one can find methods for manipulating localized Majorana wave packets. We study the fate of such flat bands when part of the surface is subjected to an exchange field induced by a ferromagnetic insulator. We use exact diagonalization to find the eigenstates and eigenenergies of the Bogoliubov-de Gennes Hamiltonian of a model system, for which an exchange field is applied along a strip on the surface of a slab. We consider different orientations of the strip and the applied field. If the spin polarization of the field-free system along the field direction is sufficiently large perturbation theory predicts that energies of states which are mostly localized on the exchange-field strip are shifted away from zero energy by an amount proportional to the field strength. On the other hand, energies corresponding to states localized on the field-free strip are only weakly affected by the field. Exact diagonalization confirms this. Moreover, we discuss a setup with a small exchange field applied to the previously field-free strip with the goal of introducing a linear dispersion. By switching this dispersion on and off, a wave packet could be moved in a certain direction. We find that in our model system, a linear dispersion can indeed be achieved. The qualitative features of this dispersion can be predicted from the momentum-dependent spin polarization of the field-free surface., Comment: 14 pages, 8 figures