Semi-Automated Workflow for Segmenting and Quantifying Ribosomes in Electron Tomograms of High-Pressure Frozen and Resin-Embedded Samples
Ribosomes are the central molecular machines of protein synthesis, and accurate counts of ribosome numbers are essential for linking translational capacity to cellular physiology. Common approaches infer ribosome abundance indirectly, for example, by measuring rRNA or ribosomal proteins, which can obscure spatial heterogeneity and sample-specific differences. This protocol presents an accessible image analysis pipeline for segmenting and counting individual ribosomes in three-dimensional transmission electron microscopy (3D-TEM) data. Established experimental procedures, namely high-pressure freezing (HPF), freeze substitution, resin embedding, serial semi-thick sectioning, and dual-axis tomography, are summarized here with direct references to primary protocols. The core of this work details electron tomography reconstruction and dual-axis stitching in IMOD 5.1, region-of-interest cropping in Fiji, and a semi-automated segmentation pipeline in arivis Pro 4.5. This pipeline inverts and denoises tomograms, detects ribosomal particles with a tuned Blob Finder routine, applies empirically derived voxel-count thresholds alongside a laboratory-specific shrinkage correction, and converts segmented counts to ribosomes per cubic micrometer and, for cultured cells, per cell. Validation shows that tomographic counts agree with independent biochemical rRNA measurements to within the same order of magnitude, and the protocol states the false-positive and false-negative limitations inherent to nonspecific heavy-metal staining.
Predicting Proteomics Data From Transcriptomics Data With MintOmics
Paired transcriptomic and proteomic profiling is a central approach in systems biology for understanding how gene-expression programs relate to protein abundance across biological conditions. In recent years, a variety of computational methods have been developed to analyze such data, including correlation-based analyses, regression models, partial least-squares approaches, and matrix-factorization techniques, while more recent deep-learning methods aim to capture nonlinear cross-omic relationships but typically require large numbers of paired samples. In these studies, the dimensionality of each assay may be large, yet the number of paired transcriptomic and proteomic observations (samples) is often small because every condition requires coordinated sample collection, sequencing, mass spectrometry, and downstream processing. This imbalance creates a challenging learning problem: a model must estimate cross-omic relationships from only a few paired samples. MintOmics addresses this limited-sample setting through masked transcriptome-to-proteome inference. The workflow pairs biological samples, filters and normalizes bulk RNA-seq and LC-MS/MS data, maps gene symbols to protein accessions, partitions biological units to prevent information leakage, and generates randomly masked computational views within each partition. A transformer is then trained to predict the abundance classes of proteins corresponding to randomly masked transcripts from the remaining unmasked transcriptomic context, followed by held-out inference, attention interpretation, and downstream enrichment analysis. Tested on the transcriptomics and proteomics data of the mouse oviduct, this approach enables the prediction of protein abundance states from transcriptomic profiles when paired multi-omics observations are limited. Its attention-based outputs provide an interpretable view of the transcriptomic features (e.g., key transcription factors) contributing to protein abundance predictions, thereby supporting downstream analyses of biological processes and regulatory mechanisms.
A Simplified Workflow for 3D Segmentation of Mouse Embryo Nuclei Using Biom3D
Three-dimensional segmentation of cell nuclei is a fundamental step in quantitative studies of morphogenesis and early embryonic development, enabling the analysis of cellular organization and behavior in complex biological tissues. However, accurate segmentation of densely packed nuclei in large volumetric microscopy datasets remains technically challenging, as touching or closely apposed nuclei are frequently merged by instance-based segmentation approaches. Several deep learning–based approaches have been developed to address this problem, including convolutional neural network architectures and shape-based segmentation methods. Although these techniques have improved segmentation performance, their adoption is often limited by complex software installation procedures, demanding dataset annotation requirements and computational expertise. Here, we present a user-oriented workflow for 3D segmentation of nuclei in mouse embryos designed to simplify and speed up image annotation, requiring little manual intervention. The protocol combines a dual-channel imaging strategy, in which a high-resolution channel optimized for nuclear boundary detection is used to generate ground-truth annotations, while a lower-resolution DAPI channel serves as the training input. Rather than Cellpose, which is difficult to train and slow in 3D, or StarDist, which is difficult to use without programming skills, we suggest training Biom3d in its default nnCore version. The workflow includes semi-automated generation of training datasets, model training and evaluation through a preconfigured Biom3d environment, and application of the trained model to new volumetric datasets. Automated preprocessing, prediction, and post-processing steps are integrated into the pipeline to facilitate high-throughput analysis. The trained model was evaluated with a Dice score of 0.81. This protocol provides an accessible and reproducible solution for non-specialist users by minimizing manual annotation efforts and eliminating most software configuration requirements. The combination of dual-channel training data generation and a preconfigured deep learning framework enables scalable and robust 3D nuclear segmentation with limited computational expertise.
A Step-by-Step UniDesign Protocol for the Fully Computational Design of PAM-Relaxed CRISPR–Cas Point-Mutation Variants
Engineering CRISPR nucleases with expanded protospacer-adjacent motif (PAM) compatibility can increase the number of genomic sites accessible to genome editing. This protocol describes a fully computational UniDesign workflow for engineering point-mutant Staphylococcus aureus Cas9 (SaCas9) variants with relaxed PAM recognition. Starting from the SaCas9–sgRNA–DNA complex structure, the protocol provides step-by-step instructions for installing and configuring UniDesign, preparing PAM-altered structural models, selecting mutable and repackable residues, constructing RESFILEs, performing three iterative design steps, and analyzing the resulting variants. The first step identifies substitutions that reduce energetic discrimination among four PAM models, yielding R1015H as the fixed background for subsequent design. The second step searches for compensatory mutations that restore protein–DNA binding strength through favorable, sequence-independent interactions. The third step combines promising substitutions to generate and prioritize triple mutants based on mean PAM-binding energy, mean absolute deviation, and structural plausibility. Designed for users with limited computational experience, this protocol enables reproduction of the SaCas9 PAM-relaxation workflow and provides an adaptable framework for engineering other protein–nucleic acid interactions.
Computational Approach for the Reliable Inference of Task States Encoded by Neurons Using Calcium Imaging
Calcium imaging has become a widely used technique for measuring the activity of identified neurons during behavior. A key goal in such investigations is to determine whether, and to what extent, neurons preferentially encode one task-relevant state over another—a property referred to as neuronal selectivity. Despite the widespread use of calcium imaging for this purpose, no consensus exists on the parameter choices required to assess neuronal selectivity reliably: choices regarding neural data representation, behavioral position tracking, temporal binning, data balancing, and statistical shuffling all vary substantially across published studies. We recently developed a systematic computational approach to address this gap. Here, we describe the step-by-step protocol for implementing this computational pipeline. Starting from preprocessed calcium imaging data, the protocol guides users through all stages of the analysis: loading and aligning neural and behavioral data, deriving four metrics of neural activity, computing a selectivity index (SI), generating null distributions using four shuffling strategies, and assigning selectivity labels to individual neurons. The protocol introduces a principled, data-driven framework for evaluating the quality of any combination of parameter choices using quantitative metrics of accuracy and robustness and for identifying the optimal combination. The framework is validated using bootstrap resampling and hold-out cross-validation, and its generality is demonstrated by application to a different, regression-based approach for assessing selectivity of individual neurons. Finally, we describe the natural extension of this framework to inferences about task states encoded by populations of neurons. Thus, our protocol provides a reusable, extensible approach to identify optimal parameter settings for reliably characterizing neural encoding of two discrete task states.
A Batch-Processing Pipeline for RNA-seq Gene Abundance Estimation
RNA sequencing (RNA-seq) datasets provide valuable opportunities to investigate gene and transcript abundance across species, tissues, developmental stages, and experimental conditions. However, processing multiple datasets consistently remains challenging because sequencing runs may differ in library layout, read length, sequencing depth, metadata quality, and analytical settings. Existing RNA-seq workflows often depend on dedicated workflow managers, substantial computational infrastructure, or advanced bioinformatics expertise, which may limit their accessibility for routine analyses. We present a modular Bash- and Python-based batch-processing pipeline for estimating gene and isoform abundance and generating expression matrices from multiple RNA-seq runs. Using an SRA RunTable, a reference genome, and its corresponding gene annotation, the workflow automates reference preparation, sequencing-data retrieval, quality assessment, read preprocessing, alignment, abundance estimation, and post-processing. It produces gene-level TPM and FPKM matrices, individual gene- and isoform-level RSEM outputs, quality-control summaries, and integrated MultiQC reports. Configurable computational resources, sample-level status tracking, automatic download retries, selective reprocessing of failed samples, and controlled removal of intermediate files allow interrupted analyses to resume without repeating completed runs while reducing storage requirements. By combining batch processing, transparent configuration, and restartable execution in a lightweight workflow, this protocol provides an accessible approach for standardized RNA-seq abundance estimation.
Quantification of Asymmetric Division in Cancer Cells via a Flow Cytometry and Cell Sorting–Based Protocol
Asymmetric partitioning of cellular components during division contributes to generating phenotypic heterogeneity and plasticity in genetically identical cells, like bacteria or cancer cells. Traditionally, this partitioning noise has been investigated using live-cell time-lapse confocal microscopy; however, while imaging offers high spatial resolution, it suffers from low statistical throughput and labor-intensive cell tracking. To overcome these limitations, we present a protocol that combines high-throughput flow cytometry and sorting with a theoretical model to precisely quantify partitioning noise during cell division. This method details the experimental setup for measuring cellular components in dividing populations, the data acquisition pipeline, and the implementation of the mathematical framework to extract inheritance parameters. Applicable to both suspension and adherent cells, we have validated our approach against live-cell imaging of adherent colon cancer cells. Hence, this flow cytometry–based approach offers a highly scalable, accessible, and statistically robust alternative to traditional imaging. It is widely applicable across diverse cell lines and biological systems, significantly advancing our capacity to study partitioning-driven phenotypic heterogeneity.
A Python-Based Workflow for Image-Based Single-Cell Phenotypic Profiling From Fluorescence Microscopy Images
While advances in omics technologies have greatly improved our understanding of cellular heterogeneity, there is an increasing need for complementary approaches that capture the spatial organization and structural dynamics of cells. Image-based single-cell phenotypic profiling provides quantitative information on cell morphology and organelle organization, offering valuable insights into cellular function and regulation. Although numerous image analysis tools are available, establishing a complete analysis workflow, from image preprocessing and segmentation to feature extraction and multivariate analysis, often requires substantial computational expertise and software integration. Here, we describe a Python-based workflow for image-based single-cell phenotypic profiling from immunofluorescence microscopy images and provide a detailed protocol for its implementation. Using synchronized HeLa cells with drug-induced mitotic spindle defects as an example, the workflow covers image loading, cell segmentation, quantitative feature extraction, profile integration, dimensionality reduction, clustering, and data visualization. The protocol is accompanied by example datasets, annotated Jupyter Notebooks, and instructions for execution in either a local Python environment or Google Colab, facilitating straightforward implementation and customization. By integrating the entire analysis pipeline within a single coding environment, this workflow enables reproducible and accessible single-cell morphological profiling without requiring specialized imaging equipment or extensive programming expertise. The workflow therefore provides a practical platform for studying cell morphology, organelle organization, and cellular dynamics across a broad range of biological applications.
Serial Cryosectioning for the Spatial Transcriptomics of Plant Tissues
Spatial transcriptomics enables genome-wide gene expression profiling while preserving tissue architecture, making it a powerful approach for studying plant developmental transitions. However, preparing small and structurally complex plant tissues for spatial transcriptomics remains technically challenging because samples must be rapidly preserved, precisely oriented, serially sectioned, and accurately positioned within the limited capture area of the Visium slides. Here, we describe an optimized workflow for cryo-embedding, serial cryosectioning, section placement, and data analysis of small plant samples for 10x Genomics Visium spatial transcriptomics. Using maize seedling shoot apices as the target, this protocol includes preparation of custom molds for optimal cutting temperature embedding, rapid fresh sample embedding, serial cryosectioning, section-position marking for Visium HD workflows, and morphological quality assessment of replicate tissue slides before transcript capture. The associated data analysis workflow includes Space Ranger processing, Seurat-based normalization and Harmony integration, anatomical domain annotation, pseudobulk and developmental trend analyses, RNA velocity, pseudotime analysis, transcription factor network analysis, single-cell reference mapping, and 3D transcriptome reconstruction. This computational workflow was developed and tested using maize Visium V1 data, but not Visium HD data. This protocol was used to generate serial spatial transcriptomes of maize shoot apices and developing leaf primordia, enabling reconstruction of gene expression transitions from the shoot apical meristem to sequential leaf developmental stages. The approach is also applicable to other small plant tissues, including Arabidopsis first true leaves and Marchantia thalli.
PIC-RNA-seq for Region-Specific Transcriptomic Analysis of Chicken Limb Bud Progenitors
Region-specific RNA sequencing is a powerful approach for investigating tissue differentiation and dynamic changes in gene expression during embryonic development. The chicken embryo has long served as an important model system in developmental biology. However, the limited availability of tissue-specific reporter lines makes region-specific RNA-seq approaches particularly valuable in this organism. Here, we applied photo-isolation chemistry-based RNA sequencing (PIC-RNA-seq) to the somatic lateral plate mesoderm (sLPM) before the emergence of limb bud progenitor cells (LPCs) and to early LPCs in chicken embryos. These analyses revealed the upregulation of multiple genes, including Hox genes, in LPCs, suggesting the initiation of their regional patterning program. This workflow enables visualization of dynamic changes in the gene expression profile of LPCs and should also be applicable to other tissues in avian embryos.
Quantitative Colocalization Analysis in Fluorescence Microscopy
Spatial organization of macromolecules is fundamental to cellular function, with colocalization providing key insights into molecular interactions and biological processes. However, quantification remains challenging due to diverse localization patterns and irregular sample geometries. Here, we present a protocol for analyzing colocalization between two fluorescent probes using coAnalyzer, a MATLAB-based software package. coAnalyzer features a user-friendly graphical interface and supports region of interest (ROI) selection, image merging, line scanning, signal isolation, scatterplot generation, and quantitative analysis. coAnalyzer enables colocalization analysis of any two fluorophores, regardless of the proteins or dyes involved. Its broad applicability across a wide range of organisms and sample types demonstrates the robustness, flexibility, and versatility of the platform.
Tapenade: Spatial Quantification of Mechanical and Genetic Fields in Dense 3D Organoids From Cell to Tissue Scale
Whole-mount 3D imaging of multilayered biological tissues enables quantitative analysis of cell states and organization in their spatial context. However, extracting unbiased and meaningful quantitative information from dense, multilayered samples remains challenging due to imaging artifacts, increased density, and limited signal-to-noise ratio. Open source bioimage analysis workflows tailored to this type of analysis are scarce, and analysis bottlenecks like image curation or cell segmentation are seldom available without coding expertise. Here, we present a step-by-step computational protocol for the analysis of dense 3D organoid datasets using the Tapenade (Thorough Analysis PipEliNe for Advanced DEep imaging) workflow. Starting from multichannel image stacks, the protocol guides users through software installation, registration and fusion of multi-view datasets, preprocessing, and nuclei segmentation. It further details the generation of quantitative outputs, including morphometric measurements, deformation fields, and spatial correlation analyses. The workflow can be executed through open-source Python scripts or user-friendly Napari interfaces, allowing interactive parameter tuning and 3D visualization at each stage. This pipeline provides an accessible and modular framework for nonspecialist users to perform reproducible, multiscale quantitative analysis of 3D organoid images, while retaining flexibility for advanced users to customize individual steps.
Digital Quantification of Membrane DAB Immunohistochemical Staining in FFPE Cervical Cancer Tissues Using an Open-Source CellProfiler Pipeline
Immunohistochemistry (IHC) is a highly specific and widely used laboratory technique for assessing protein localization and expression in tissue samples. Interpretation of 3,3’ diamino benzidine (DAB)-based IHC is often based on observer-dependent manual scoring or traditional imaging software, which may show variability in DAB staining quantification. Furthermore, conventional image analysis tools often face limitations in precisely defining cell boundaries and quantifying membrane-specific signals. In this study, we present a standardized image analysis workflow using CellProfiler, an open-source software for image analysis for the quantification of membrane staining intensity in IHC images captured from slides prepared using formalin-fixed paraffin-embedded (FFPE) human cervical cancer tissue sections. The image analysis workflow was demonstrated using ASCT2 (SLC1A5), a membrane-localized amino acid transporter, as a representative biomarker for membrane-associated protein expression. This protocol involves image preprocessing, object identification, segmentation, and intensity measurement modules to distinguish cell membranes from cytoplasmic regions, enabling automated quantification of membrane intensity signals. The CellProfiler pipeline demonstrated improved accuracy in cell boundary identification and quantification of membrane-specific staining intensity. This is a rapid quantification process, since processing of each image only takes a few seconds; therefore, the analysis for 100 images can be performed within 10–15 min. This segmentation and quantification strategy is applicable to other membrane-based biomarkers after appropriate optimization of segmentation parameters. Following further minor modifications to the object identification modules, this pipeline can be used to detect and quantify cytoplasm- or nuclei-localized DAB-IHC markers across different tissue types. Overall, this protocol provides a standardized, user-friendly, and reproducible workflow for quantitative IHC image analysis that can be broadly applied to the study of protein biomarkers of different localizations, such as nuclei, cytoplasm, and cell membranes from different tissue types.
Humanizing Antibodies and Nanobodies From Scratch With HuDiff
Antibody (Ab) and nanobody (Nb) humanization is essential for reducing immunogenicity in therapeutic applications. HuDiff is an adaptive autoregressive diffusion approach that generates humanized antibodies and nanobodies from scratch using only complementarity-determining region sequences as input, eliminating the need for preexisting human templates. The method follows a two-stage training pipeline: pretraining on human antibody sequences to learn framework region patterns, followed by fine-tuning on target-species sequences. HuDiff-Ab processes paired heavy and light chains for conventional antibodies, while HuDiff-Nb can incorporate a specialized inpainting mode to preserve critical nanobody framework residues. This protocol provides a complete step-by-step guide for implementing HuDiff, covering data preparation, model training, and sequence generation.