Biological Findings Research Articles

Formaldehyde Fixation Helps Preserve the Proteome State during Single-Cell Proteomics Sample Processing and Analysis.

Mass spectrometry-based single-cell proteomics (SCP) is gaining momentum but remains limited to a few laboratories due to the high costs and specialized expertise required. The ability to send samples to specialized core facilities would benefit nonspecialist laboratories and popularize SCP for biological applications. However, no methods have been tested in SCP to "freeze" the proteome state while maintaining cell integrity for transfer between laboratories or prolonged sorting using fluorescence-activated cell sorting (FACS). This study evaluates whether short-term formaldehyde (FA) fixation can maintain the cell states. We demonstrate that short-term FA fixation does not substantially affect protein recovery, even without heating and strong detergents, and maintains analytical depth compared with classical workflows. Fixation also preserves drug-induced specific perturbations of the protein abundance during cell sorting and sample preparation for SCP analysis. Our findings suggest that FA fixation can facilitate SCP by enabling sample shipping and prolonged sorting, potentially democratizing access to SCP technology and expanding its application in biological research, thereby accelerating discoveries in cell biology and personalized medicine.

KaMLs for Predicting Protein pKa Values and Ionization States: Are Trees All You Need?

Despite its importance in understanding biology and computer-aided drug discovery, the accurate prediction of protein ionization states remains a formidable challenge. Physics-based approaches struggle to capture the small, competing contributions in the complex protein environment, while machine learning (ML) is hampered by the scarcity of experimental data. Here, we report the development of pKa ML (KaML) models based on decision trees and graph attention networks (GAT), exploiting physicochemical understanding and a new experiment pKa database (PKAD-3) enriched with highly shifted pKa's. KaML-CBtree significantly outperforms the current state of the art in predicting pKa values and ionization states across all six titratable amino acids, notably achieving accurate predictions for deprotonated cysteines and lysines─a blind spot in previous models. The superior performance of KaMLs is achieved in part through several innovations, including the separate treatment of acid and base, data augmentation using AlphaFold structures, and model pretraining on a theoretical pKa database. We also introduce the classification of protonation states as a metric for evaluating pKa prediction models. A meta-feature analysis suggests a possible reason for the lightweight tree model to outperform the more complex deep learning GAT. We release an end-to-end pKa predictor based on KaML-CBtree and the new PKAD-3 database, which facilitates a variety of applications and provides the foundation for further advances in protein electrostatic research.

KaMLs for Predicting Protein pKa Values and Ionization States: Are Trees All You Need?

Despite its importance in understanding biology and computer-aided drug discovery, the accurate prediction of protein ionization states remains a formidable challenge. Physicsbased approaches struggle to capture the small, competing contributions in the complex protein environment, while machine learning (ML) is hampered by scarcity of experimental data. Here we report the development of pKa ML (KaML) models based on decision trees and graph attention networks (GAT), exploiting physicochemical understanding and a new experiment pKa database (PKAD-3) enriched with highly shifted pKa's. KaML-CBtree significantly outperforms the current state of the art in predicting pKa values and ionization states across all six titratable amino acids, notably achieving accurate predictions for deprotonated cysteines and lysines - a blind spot in previous models. The superior performance of KaMLs is achieved in part through several innovations, including separate treatment of acid and base, data augmentation using AlphaFold structures, and model pretraining on a theoretical pKa database. We also introduce the classification of protonation states as a metric for evaluating pKa prediction models. A meta-feature analysis suggests a possible reason for the lightweight tree model to outperform the more complex deep learning GAT. We release an end-to-end pKa predictor based on KaML-CBtree and the new PKAD-3 database, which facilitates a variety of applications and provides the foundation for further advances in protein electrostatics research.

Fourier or Wavelet bases as counterpart self-attention in spikformer for efficient visual classification

Energy-efficient spikformer has been proposed by integrating the biologically plausible spiking neural network (SNN) and artificial transformer, whereby the spiking self-attention (SSA) is used to achieve both higher accuracy and lower computational cost. However, it seems that self-attention is not always necessary, especially in sparse spike-form calculation manners. In this article, we innovatively replace vanilla SSA (using dynamic bases calculating from Query and Key) with spike-form Fourier transform, wavelet transform, and their combinations (using fixed triangular or wavelets bases), based on a key hypothesis that both of them use a set of basis functions for information transformation. Hence, the Fourier-or-Wavelet-based spikformer (FWformer) is proposed and verified in visual classification tasks, including both static image and event-based video datasets. The FWformer can achieve comparable or even higher accuracies (0.4%–1.5%), higher running speed (9%–51% for training and 19%–70% for inference), reduced theoretical energy consumption (20%–25%), and reduced graphic processing unit (GPU) memory usage (4%–26%), compared to the standard spikformer. Our result indicates the continuous refinement of new transformers that are inspired either by biological discovery (spike-form), or information theory (Fourier or Wavelet transform), is promising.

Mechanistic insights and approaches for beta cell regeneration.

Diabetes is characterized by variable loss of insulin-producing beta cells, and new regenerative approaches to increasing the functional beta cell mass of patients hold promise for reversing disease progression. In this Review, we summarize recent chemical biology breakthroughs advancing our knowledge of beta cell regeneration. We present current chemical-based tools, sensors and mechanistic insights into pathways that can be targeted to enhance beta cell regeneration in model organisms. We group the pathways according to the cellular processes they affect, that is, proliferation, conversion of other mature cell types to beta cells and beta cell differentiation from progenitor-like populations. We also suggest assays for assessing the functionality of the regenerated beta cells. Although regeneration processes differ between animal models, such as zebrafish, mice and pigs, regenerative mechanisms identified in any one animal model may be translatable to humans. Overall, chemical biology-based approaches in beta cell regeneration give hope that specific molecular pathways can be targeted to enhance beta cell regeneration.

MsiFlow: automated workflows for reproducible and scalable multimodal mass spectrometry imaging and microscopy data analysis

Multimodal imaging by matrix-assisted laser desorption ionisation mass spectrometry imaging (MALDI MSI) and microscopy holds potential for understanding pathological mechanisms by mapping molecular signatures from the tissue microenvironment to specific cell populations. However, existing software solutions for MALDI MSI data analysis are incomplete, require programming skills and contain laborious manual steps, hindering broadly applicable, reproducible, and high-throughput analysis to generate impactful biological discoveries. Here, we present msiFlow, an accessible open-source, platform-independent and vendor-neutral software for end-to-end, high-throughput, transparent and reproducible analysis of multimodal imaging data. msiFlow integrates all necessary steps from raw data import to analytical visualisation along with state-of-the-art and self-developed algorithms into automated workflows. Using msiFlow, we unravel the molecular heterogeneity of leukocytes in infected tissues by spatial regulation of ether-linked phospholipids containing arachidonic acid. We anticipate that msiFlow will facilitate the broad applicability of MSI in multimodal imaging to uncover context-dependent cellular regulations in disease states.

A Novel Tumor on Chip Mimicking the Breast Cancer Microenvironment for Dynamic Drug Screening

In light of the emerging breakthroughs in cancer biology, drug discovery, and personalized medicine, Tumor-on-Chip (ToC) platforms have become pivotal tools in current biomedical research. This study introduced a novel rapid prototyping approach for the fabrication of a ToC device using laser-patterned poly(methyl methacrylate) (PMMA) layers integrated with a polylactic acid (PLA) electrospun scaffold, enabling dynamic drug delivery and the assessment of therapeutic efficacy in cancer cells. Traditional drug screening methods, such as conventional cell cultures, mimic certain aspects of cancer progression but fail to capture critical features of the tumor microenvironment (TME). While animal models offer a closer approximation of tumor complexity, they are limited in their ability to predict human drug responses. Here, we evaluated the ability of our ToC device to recapitulate the interactions between cancer and TME cells and its efficacy in evaluating the drug response of breast cancer cells. The functional design of the proposed ToC system offered substantial potential for a wide range of applications in cancer research, significantly accelerating the preclinical assessment of new therapeutic agents.

Multifaceted Actions of Neurosteroids.

Neurosteroids modulate neuronal function and are promising therapeutic agents for neuropsychiatric disorders. Neurosteroid analogues are approved for treating postpartum depression and are of interest in other disorders. GABA-A receptors are well characterized targets of natural neurosteroids, but other biological pathways are likely relevant to therapeutic mechanisms and/or to off-target effects. We performed hypothesis-generating in silico analyses and broad in vitro biological screens to assess the range of actions of neurosteroids analogues of varying structural attributes. We employed in silico molecular similarity analysis and network pharmacology to elucidate likely targets. This analysis confirmed likely targets beyond GABA-A receptors. We then functionally screened 19 distinct neurosteroid structures across 78 targets representing interconnected signaling pathways, complemented with a limited screen of kinase activation. Results revealed unanticipated modulation of targets by neurosteroids with some structural selectivity. Many compounds-initiated androgen receptor translocation with little or no enantioselectivity. Modulation of multiple G-protein receptors was also unexpected. Neurosteroids are ascendant treatments in neuropsychiatry, but their full spectrum of actions remains unclear. This virtual and biological screening discovery approach opens new vistas for exploring mechanism of neurosteroids analogues. The multifaceted approach provides an unbiased, holistic exploration of the potential effects of neurosteroids across various molecular targets and provides a platform for future validation studies to aid drug discovery.

Deep learning methods for proteome-scale interaction prediction.

Proteome-scale interaction prediction is essential for understanding protein functions and disease mechanisms. Traditional experimental methods are often limited by scale and complexity, driving the need for computational approaches. Deep learning has emerged as a powerful tool, enabling high-throughput, accurate predictions of protein interactions. This review highlights recent advances in deep learning methods for protein-protein and protein-ligand interaction screening, along with datasets used for model training. Despite the progress with deep learning, challenges such as data quality and validation biases remain. We also discuss the increasing importance of integrating structural information to enhance prediction accuracy and how structure-based deep learning approaches can help overcome current limitations, ultimately advancing biological research and drug discovery.

Modular Synthesis of Planar-Chiral Cyclononenes via trans-Retentive Trapping of π-Allyl-Pd Dipoles.

Trans-cycloalkenes are abundant in bioactive natural products and have been used as powerful tools in chemical biology and drug discovery. However, strategies for the modular synthesis of trans-cycloalkenes, especially planar-chiral medium-sized ones, with high efficiency and selectivity, still remain elusive. Herein, we report a Pd-catalyzed asymmetric [7 + 2] cyclization strategy to address this challenge. As a result, two types of planar-chiral trans-cyclononenes bearing all-carbon chiral quaternary stereocenters are produced in up to 85% yield, >99% ee, and >19:1 dr (42 examples). The key to this success is the maintenance of the trans-2H configuration of the π-allyl-Pd species along with unusual linear selectivity during the H-bond-assisted cyclization process. In addition, the conversion of planar chirality to central chirality and its application in selective bioimaging of cancer cells via a bioorthogonal reaction were performed to demonstrate the synthetic value of this methodology.

Sherlock-Genome: an R Shiny application for genomic analysis and visualization

MotivationNext-generation sequencing technologies, such as whole genome sequencing (WGS), have become prominent in cancer genomics. However, managing, visualizing, and integratively analyzing WGS results across various bioinformatic pipelines remains challenging, particularly for non-bioinformaticians, hindering the usability of WGS data for biological discovery.ResultsWe developed Sherlock-Genome, an R Shiny app for data harmonization, visualization, and integrative analysis of WGS-based cancer genomics studies. Following FAIR principles, Sherlock-Genome provides a platform and guidelines for managing and sharing finalized sample-level WGS analysis results, enabling users to upload results, inspect analyses locally, and perform integrative analyses. It includes modules for major cancer genomic analyses, allowing interactive data visualizations and integrative analyses with other data types. Sherlock-Genome supports both local and cloud deployment, facilitating the sharing of results for related publications. This tool has the potential to be widely adopted in cancer genomics, significantly enhancing the accessibility and usability of sample-level WGS analysis results for comprehensive biological discovery and research advancements.Availability and implementationThe source code and installation instructions for Sherlock-Genome can be accessed via Github https://github.com/xtmgah/Sherlock-Genome. Documentation and data requirements for user project data can also be found on the same GitHub page.

Geroscience in heart failure: the search for therapeutic targets in the shared pathobiology of human aging and heart failure

Age is a major risk factor for heart failure, but one that has been historically viewed as non-modifiable. Emerging evidence suggests that the biology of aging is malleable, and can potentially be intervened upon to treat age-associated chronic diseases, such as heart failure. While aging biology represents a new frontier for therapeutic target discovery in heart failure, the challenges of translating Geroscience research to the clinic are multifold. In this review, we propose a strategy that prioritizes initial target discovery in human biology. We review the rationale for starting with human omics, which has generated important insights into the shared (patho)biology of human aging and heart failure. We then discuss how this knowledge can be leveraged to identify the mechanisms of aging biology most relevant to heart failure. Lastly, we provide examples of how this human-first Geroscience approach, when paired with rigorous functional assessments in preclinical models, is leading to early-stage clinical development of gerotherapeutic approaches for heart failure.

GENA-LM: a family of open-source foundational DNA language models for long sequences.

Recent advancements in genomics, propelled by artificial intelligence, have unlocked unprecedented capabilities in interpreting genomic sequences, mitigating the need for exhaustive experimental analysis of complex, intertwined molecular processes inherent in DNA function. A significant challenge, however, resides in accurately decoding genomic sequences, which inherently involves comprehending rich contextual information dispersed across thousands of nucleotides. To address this need, we introduce GENA language model (GENA-LM), a suite of transformer-based foundational DNA language models capable of handling input lengths up to 36 000 base pairs. Notably, integrating the newly developed recurrent memory mechanism allows these models to process even larger DNA segments. We provide pre-trained versions of GENA-LM, including multispecies and taxon-specific models, demonstrating their capability for fine-tuning and addressing a spectrum of complex biological tasks with modest computational demands. While language models have already achieved significant breakthroughs in protein biology, GENA-LM showcases a similarly promising potential for reshaping the landscape of genomics and multi-omics data analysis. All models are publicly available on GitHub (https://github.com/AIRI-Institute/GENA_LM) and on HuggingFace (https://huggingface.co/AIRI-Institute). In addition, we provide a web service (https://dnalm.airi.net/) allowing user-friendly DNA annotation with GENA-LM models.

The Influence of Phosphoinositide Lipids in the Molecular Biology of Membrane Proteins: Recent Insights from Simulations.

The phosphoinositide family of membrane lipids play diverse and critical roles in eukaryotic molecular biology. Much of this biological activity derives from interactions of phosphoinositide lipids with integral and peripheral membrane proteins, leading to modulation of protein structure, function, and cellular distribution. Since the discovery of phosphoinositides in the 1940s, combined molecular biology, biophysical, and structural approaches have made enormous progress in untangling this vast and diverse cellular network of interactions. More recently, in silico approaches such as molecular dynamics simulations have proven to be an asset in prospectively identifying, characterising, explaining the structural basis of these interactions, and in the best cases providing atomic level testable hypotheses on how such interactions control the function of a given membrane protein. This review details a number of recent seminal discoveries in phosphoinositide biology, enabled by advanced biomolecular simulation, and its integration with molecular biology, biophysical, and structural biology approaches. The results of the simulation studies agree well with experimental work, and in a number of notable cases have arrived at the key conclusion several years in advance of the experimental structures. SUMMARY: Hedger and Yen review developments in simulations of phosphoinositides and membrane proteins.

Genome-wide association studies in a large Korean cohort identify novel quantitative trait loci for 36 traits and illuminate their genetic architectures.

Genome-wide association studies (GWAS) have been predominantly conducted in populations of European ancestry, limiting opportunities for biological discovery in diverse populations. We report GWAS findings from 153,950 individuals across 36 quantitative traits in the Korean Cancer Prevention Study-II (KCPS2) Biobank. We discovered 301 novel genetic loci in KCPS2, including an association between thyroid-stimulating hormone and CD36. Meta-analysis with the Korean Genome and Epidemiology Study, Biobank Japan, Taiwan Biobank, and UK Biobank identified 4,588 loci that were not significant in any contributing GWAS. We describe differences in genetic architectures across these East Asian and European samples. We also highlight East Asian specific associations, including a known pleiotropic missense variant in ALDH2, which fine-mapping identified as a likely causal variant for a diverse set of traits. Our findings provide insights into the genetic architecture of complex traits in East Asian populations and highlight how broadening the population diversity of GWAS samples can aid discovery.

Harnessing the AI/ML in Drug and Biological Products Discovery and Development: The Regulatory Perspective.

Artificial Intelligence (AI) has the disruptive potential to transform patients' lives via innovations in pharmaceutical sciences, drug development, clinical trials, and manufacturing. However, it presents significant challenges, ethical concerns, and risks across sectors and societies. AI's rapid advancement has revealed regulatory gaps as existing public policies struggle to keep pace with the challenges posed by these emerging technologies. The term AI itself has become commonplace to argue that greater "human oversight" for "machine intelligence" is needed to harness the power of this revolutionary technology for both potential and risk management, and hence to call for more practical regulatory guidelines, harmonized frameworks, and effective policies to ensure safety, scalability, data privacy, and governance, transparency, and equitable treatment. In this review paper, we employ a holistic multidisciplinary lens to survey the current regulatory landscape with a synopsis of the FDA workshop perspectives on the use of AI in drug and biological product development. We discuss the promises of responsible data-driven AI, challenges and related practices adopted to overcome limitations, and our practical reflections on regulatory oversight. Finally, the paper outlines a path forward and future opportunities for lawful ethical AI. This review highlights the importance of risk-based regulatory oversight, including diverging regulatory views in the field, in reaching a consensus.

PhiSiCal-Checkup: A Bayesian framework to validate amino acid conformations within experimental protein structures

As structural biology and drug discovery depend on high-quality protein structures, assessment tools are essential. We describe a new method for validating amino-acid conformations: "PhiSiCal ([Formula: see text]al) Checkup." Twenty new joint probability distributions in the form of statistical mixture models explain the empirical distributions of dihedral angles [Formula: see text] of canonical amino acids in experimental protein structures. Marginal and conditional probability distributions for subsets of dihedral angles are derived from these joint mixture models. Together, these distributions are employed to measure rapidly the information-theoretic "favorability" of any proposed experimental protein structure. The inferred statistical models and measures overcome several shortcomings and afford improvements over the current state of the art in amino-acid conformation verification. Experimental comparisons are made against current protein conformation verification software. In a number of examples, we pick up outliers that are invisible to current methods. We also calculate, as part of verification, the sensitivity of favorability to small changes in a proposed structure accounting for the precision of coordinates. In some cases a near neighbor of a proposed amino-acid conformation may be either less or more favorable. This raises the question, is the current reliance on fixed "thresholds" for validation a good thing? PhiSiCal-Checkup is freely available for online and offline (open-source) use from https://lcb.infotech.monash.edu.au/phisical/checkup.

Data-driven model discovery and model selection for noisy biological systems.

Biological systems exhibit complex dynamics that differential equations can often adeptly represent. Ordinary differential equation models are widespread; until recently their construction has required extensive prior knowledge of the system. Machine learning methods offer alternative means of model construction: differential equation models can be learnt from data via model discovery using sparse identification of nonlinear dynamics (SINDy). However, SINDy struggles with realistic levels of biological noise and is limited in its ability to incorporate prior knowledge of the system. We propose a data-driven framework for model discovery and model selection using hybrid dynamical systems: partial models containing missing terms. Neural networks are used to approximate the unknown dynamics of a system, enabling the denoising of the data while simultaneously learning the latent dynamics. Simulations from the fitted neural network are then used to infer models using sparse regression. We show, via model selection, that model discovery using hybrid dynamical systems outperforms alternative approaches. We find it possible to infer models correctly up to high levels of biological noise of different types. We demonstrate the potential to learn models from sparse, noisy data in application to a canonical cell state transition using data derived from single-cell transcriptomics. Overall, this approach provides a practical framework for model discovery in biology in cases where data are noisy and sparse, of particular utility when the underlying biological mechanisms are partially but incompletely known.

Advances and challenges in precision imaging.

Technological innovations in genomics and related fields have facilitated large sequencing efforts, supported new biological discoveries in cancer, and spawned an era of liquid biopsy biomarkers. Despite these advances, precision oncology has practical constraints, partly related to cancer's biological diversity and spatial and temporal complexity. Advanced imaging technologies are being developed to address some of the current limitations in early detection, treatment selection and planning, drug delivery, and therapeutic response, as well as difficulties posed by drug resistance, drug toxicity, disease monitoring, and metastatic evolution. We discuss key areas of advanced imaging for improving cancer outcomes and survival. Finally, we discuss practical challenges to the broader adoption of precision imaging in the clinic and the need for a robust translational infrastructure.

Unifying biology of neurodegeneration in lysosomal storage diseases.

There are currently at least 70 characterised lysosomal storage diseases (LSD) resultant from inherited single-gene defects. Of these, at least 30 present with central nervous system (CNS) neurodegeneration and overlapping aetiology. Substrate accumulation and dysfunctional neuronal lysosomes are common denominator, but how variants in 30 different genes converge on this central cellular phenotype is unclear. Equally unresolved is how the accumulation of a diverse spectrum of substrates in the neuronal lysosomes results in remarkably similar neurodegenerative outcomes. Conversely, how is it that many other monogenic LSDs cause only visceral disease? Lysosomal substance accumulation in LSDs with CNS neurodegeneration (nLSD) includes lipofuscinoses, mucopolysaccharidoses, sphingolipidoses and glycoproteinoses. Here, we review the latest discoveries in the fundamental biology of four classes of nLSDs, comparing and contrasting new insights into disease mechanism with emerging evidence of unifying convergence.