生物信息学与计算生物学

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.

Riboswitches are structured non-coding RNA elements that regulate gene expression in response to small molecules; they serve as valuable systems in both public health and biophysical research by elucidating principles around RNA–ligand interactions, structure, and cellular function. Traditional approaches to studying riboswitches have relied on low-throughput techniques such as reporter assays or gel electrophoresis analysis of transcriptional products, which are limited in scalability. In this study, we present a high-throughput protocol to characterize the transcriptional activity of nearly 2,000 natural variants of the fluoride riboswitch in in vitro transcription. Starting with bioinformatics, we compiled a comprehensive dataset of riboswitch variants and then employed massive parallel oligonucleotide synthesis to generate an oligo pool of the riboswitch library. This pool was transcribed in vitro, converted into an Illumina-compatible next-generation sequencing (NGS) library, and analyzed to identify transcriptionally active riboswitch candidates. The workflow integrates natural riboswitch bioinformatic acquisition into a quantitative readout in a single streamlined pipeline, enabling large-scale exploration of transcriptional riboswitch function. This protocol offers a scalable method for mapping genotype-to-function relationships across transcriptional riboswitch families, accelerating the identification of functional variants for desired applications.

Accurate, sensitive quantification of B-lineage cells is critical for pharmacodynamic evaluation of B cell–targeted therapies in lupus nephritis (LN) clinical trials. While high-dimensional discovery platforms offer broad profiling, they often lack the sensitivity, quantitative rigor, and throughput needed for precise cell enumeration in renal trial needle biopsies. Traditional immunostaining is hampered by CD20-directed therapeutic masking or downregulation, inadequate sensitivity of CD19 in FFPE tissue, and confounding renal tubular CD138 expression. This protocol details an automated, fit-for-purpose, 5-plex sequential tyramide signal amplification (TSA)-based immunofluorescence assay (CD38, CD79a, CD19, Ki-67, CD138) developed on the Ventana Discovery Ultra platform for deployment on single tissue sections. The workflow anchors B-cell detection on CD79a to ensure sensitivity and utilizes CD38 as an obligate co-marker for CD138+ antibody-secreting cells (ASCs) to definitively exclude the CD138+ epithelial background. Following acquisition via fluorescence whole-slide imaging, a digital analysis pipeline utilizing InstanSeg-based automated segmentation rigorously classifies cell phenotypes to generate precise spatial densities (cells/mm²). This validated protocol maximizes data yield from scarce clinical biopsies while providing high-precision quantitative monitoring of longitudinal therapeutic depletion in the renal microenvironment.

Actin filaments undergo dynamic growth and disassembly at their ends, regulated by many actin-binding proteins. However, structural analysis of filament end dynamics has been challenging due to the low abundance of filament ends in cryo-electron microscopy (cryo-EM) micrographs, their intrinsic polymorphisms, and the diversity and flexibility of end-binding proteins. Here, we describe a standardized cryo-EM protocol for determining actin filament end structures. First, short actin filaments are generated either biochemically using capping or severing proteins or mechanically through shearing. Filaments are then vitrified under conditions optimized for each specific end-binding protein. We describe data collection parameters using a 300 kV Titan Krios G3i microscope, including optimized grid preparation and imaging settings. Finally, we present a data processing pipeline for filament end structure determination based on machine learning–based particle picking, masking, and sorting strategies. This protocol has enabled the determination of multiple high-resolution structures of free, capped, elongating, and depolymerizing actin filament ends, and we further discuss considerations for extending this approach to other end-binding proteins.

CRISPR-Cas9 barcoding technologies enable cells to record molecular events as permanent genetic changes that can be read out retrospectively. This protocol describes the implementation of a CRISPR-based recording system that gradually accumulates mutations over extended periods and is compatible with standard single-cell RNA sequencing (scRNA-seq) workflows. By temporally regulating CRISPR activity, the system generates mutational barcodes that can be captured together with individual cell transcriptomes. These barcodes are subsequently decoded using computational reconstruction approaches to infer temporal information, enabling the joint analysis of cellular states and time-resolved molecular histories. This approach provides a single-cell-compatible framework for studying dynamic biological processes in heterogeneous mouse embryonic stem cell (mESC)-derived systems, with potential extension to other biological systems.

Chemotherapy-resistant persister cells are a major driver of cancer recurrence, yet their epigenetic basis remains poorly characterized. This protocol describes a computational pipeline for identifying DNA-binding factors (DBFs) that are enriched in accessible chromatin that collectively define a persister cell signature (PCS). Starting from single-nucleus ATAC-seq (snATAC-seq) data processed through the 10x Genomics CellRanger ARC pipeline, this protocol covers (1) the creation of a Seurat/Signac object with ATAC peaks, (2) the optional integration of DNA-binding data from the ReMap2022 database as a per-cell chromatin module assay, (3) differential accessibility analysis across clinically defined comparison groups, and (4) identifying and defining the top enriched DBFs as the PCS. This approach is applicable to any snATAC-seq dataset in which cells can be grouped by clinical response, treatment status, or resistance phenotype.

Accurate quantification of plant disease severity is essential for evaluating host–pathogen interactions and assessing the effectiveness of disease management strategies. Traditional visual scoring methods and manual estimation of infected tissue are widely used but are often subjective and prone to observer bias. Digital image analysis offers an objective alternative by enabling automated identification and quantification of symptomatic plant tissues based on color and spatial characteristics.

Here, we present a MATLAB-based image processing protocol for differentiating diseased and healthy plant tissue from digital leaf images. The workflow involves acquisition of standardized leaf images, conversion of RGB images into hue-saturation-value (HSV) color space, segmentation of diseased tissue using defined HSV thresholds, refinement of the segmented mask through morphological operations, and extraction of the whole leaf area. The protocol then calculates the diseased area and total leaf area in pixels and computes the percentage of infected tissue. The method uses MATLAB together with the Image Processing Toolbox and can be implemented using simple scripts. This protocol enables rapid and reproducible quantification of disease severity in plant leaves exhibiting visually distinct symptoms such as necrotic lesions or blight patches. By minimizing observer bias and providing quantitative measurements of infected area, the protocol offers a practical and reproducible approach for plant disease phenotyping and evaluation of disease management strategies across diverse plant-pathogen systems where diseased tissues can be clearly distinguished from healthy tissues under reasonably controlled imaging conditions.

Estimation of model accuracy (EMA) is a critical step in protein structure prediction, enabling the ranking and selection of models in the absence of experimental structures. EMA methods aim to function independently of modeling approaches, ensuring broad applicability across diverse prediction workflows. Recent state-of-the-art EMA methods often improve estimation accuracy by incorporating consensus information from model pools, multiple sequence alignments (MSAs), structural templates, or protein language model representations. However, these strategies typically incur substantial computational cost or rely on information derived from the modeling process itself, which may introduce bias and compromise the independence of the assessment. This protocol describes the use of MViewEMA for global accuracy estimation of protein complex models from a single input structure. MViewEMA extracts residue–residue interaction features from complementary micro-, meso-, and macro-environmental perspectives and integrates multi-scale structural representations through a multi-view representation learning framework to predict global confidence scores. The protocol provides detailed procedures for input structure preparation, feature extraction, model inference, and global confidence score output, together with a tutorial for using the MViewEMA web server. The protocol provides a workflow based solely on structural information from the input model, achieving a balance between computational efficiency and estimation accuracy. It enables large-scale evaluation and selection of predicted models for protein structure prediction and downstream structural analysis applications.

Super-resolution fluorescence microscopy enables the visualization of protein structures at nanometer resolution, providing insights into receptor organization on the plasma membrane that are essential for the development and optimization of immunotherapies. In this context, monoclonal antibodies are employed, which typically bind only a subset of available membrane receptors, due to steric hindrance or otherwise limited epitope accessibility, to quantify the accessible targets. These accessible targets, rather than the total receptor density, are critical for determining therapeutic efficacy. Here, we present a simplified, robust protocol to quantify antibody-accessible endogenous receptors using monoclonal antibodies directly labeled with fluorescent dyes in combination with total internal reflection fluorescence (TIRF) direct stochastic optical reconstruction microscopy (dSTORM). The method employs optimized labeling and fixation conditions to preserve the native receptor distribution, enabling precise quantification of accessible receptors and their stoichiometry at single-molecule resolution. Omitting secondary antibodies and minimizing fixation-induced artifacts prevents artificial clustering and maintains the physiological binding pattern of therapeutic antibodies. The standardized workflow delivers therapy-relevant information about receptor accessibility and organization underlying therapeutic antibody binding, thereby advancing the mechanistic understanding of immunotherapy resistance and personalized treatment strategies across diverse membrane protein targets.

RNA sequencing (RNA-seq) has revolutionized transcriptomics, ribosome footprinting, and polysome profiling, providing a wealth of data. Many RNA-based omics typically remove ribosomal RNA (rRNA) or select for messenger RNA (mRNA) prior to sequencing, thereby enriching reads that map to the translationally active part of the transcriptome. Prokaryotic mRNA lacks the 3′ polyadenylated tail, which excludes the use of poly(A)-based selection methods. While commercial rRNA depletion products exist for prokaryotes, their proprietary nature and potential inefficiency with non-model organisms are factors that may limit broad-scale application. To mitigate this issue, we designed DepStep, a consolidated workflow for one-step rRNA depletion using species-specific biotinylated antisense probes for selective hybridization and removal of the target rRNA molecules. As a proof-of-concept, RNA-seq libraries of the psychrophilic gram-negative bacterium Shewanella glacialimarina TZS-4_T were prepared using both DepStep and a commercial rRNA depletion kit for gram-negative bacteria, to which DepStep was benchmarked. DepStep compares favorably to the commercial depletion kit; it removes >98.6% of the rRNA content in the sample, resulting in sequencing libraries where the coding DNA sequence (CDS) reads account for >80% of the total read count. Importantly, DepStep’s cost-per-sample is three times lower than the commercial kit, establishing DepStep as a simple yet cost-effective alternative to commercial solutions.