Bioinformatics and Computational Biology

DNA language models, such as DNABERT-2, have recently enabled the accurate prediction of functional sequence elements across species. However, the practical, protocol-style steps needed to transform these resources into training datasets, fine-tune the official DNABERT-2 model, and evaluate classifier performance have not been explicitly described. Herein, we present a step-by-step computational protocol for preparing training data, fine-tuning DNABERT-2, and evaluating sequence-level binary classifiers using readily available command-line tools. The protocol has been demonstrated using RNA off-target sites induced by cytosine base editors, detected by our PiCTURE pipeline from RNA sequencing (RNA-seq) data, and extended to core promoter prediction using the EPDnew database. We describe how to derive positive and negative sequence sets into DNABERT-2 compatible datasets, and fine-tune the official pretrained model of DNABERT-2 using the datasets. We also demonstrate how to compute the standard performance metrics and compare the model outputs with the baselines. This protocol will help researchers adapt DNA foundation models to new genomic tasks, including the safety assessment of genome editing tools and the functional annotation of regulatory sequences.

Protein–protein interactions (PPIs) govern nearly all aspects of cellular physiology, yet identifying these interactions under native conditions remains challenging. Here, we present TIE-UP-SIN (targeted interactome experiment for unknown proteins by stable isotope normalization), a robust method for in vivo identification and quantification of PPIs in bacterial systems. The protocol combines metabolic labeling with ¹⁵N isotopes, reversible formaldehyde crosslinking, affinity purification, and quantitative mass spectrometry. TIE-UP-SIN preserves transient or weak interactions during purification and quantifies interaction partners using internal light/heavy peptide ratios, reducing experimental variability. The method employs a triple-sample design to distinguish specific from nonspecific interactors and can be adapted to various bacterial species and affinity tags. Data analysis is streamlined through a user-friendly web application (https://shiny-fungene.biologie.uni-greifswald.de/TIE_UP_SIN_app) that automates statistical analysis, normalization, and visualization, requiring no programming expertise. The entire workflow from cell culture to mass spectrometry data acquisition takes approximately 4–5 days, with data analysis completed in 1–2 days using the web application.

Transcription factors (TFs) regulate gene expression by binding to cis-regulatory elements in the genome. Understanding transcriptional regulation requires genome-wide characterization of TF occupancy across different chromatin contexts, yet simultaneous assessment of TF binding for multiple factors remains technically challenging. Here, we describe a detailed and reproducible protocol for cFOOT-seq, a cytosine deaminase–based genomic footprinting assay by sequencing, which enables antibody-independent, base-resolution profiling of chromatin accessibility, nucleosome organization, and TF occupancy. In cFOOT-seq, the double-stranded DNA (dsDNA) cytosine deaminase SsdA_tox converts cytosine to uracil in accessible chromatin, whereas TF binding and nucleosome occupancy locally protect DNA from deamination. Using the FootTrack analysis framework, deamination patterns generated by cFOOT-seq are quantitatively analyzed to derive standardized footprint and chromatin organization profiles at base resolution across the genome. Because cFOOT-seq preserves genomic DNA integrity during deamination-based footprinting, it is compatible with ATAC-seq-based chromatin enrichment. ATAC-combined implementations reduce sequencing depth requirements and improve scalability for footprint-focused analyses, supporting applications in low-input and single-cell settings. This protocol provides a practical framework for genome-wide TF footprint profiling and can be readily applied to dissect gene regulatory mechanisms in development, immunity, and disease, including cancer.

Extrachromosomal circular DNA (eccDNA) is a type of circular DNA that exists independently of chromosomes and has garnered significant attention in various fields, particularly in the context of smaller eccDNAs, which have considerable roles in gene regulation through various mechanisms. Current methods such as Circle-Seq and 3SEP can enrich small eccDNAs during sample preparation, but most bioinformatics pipelines remain challenging, exhibiting low accuracy and efficiency. This protocol describes the detailed workflow of a newly developed bioinformatics analysis pipeline, named EccDNA Caller based on Consecutive Full Pass (ECCFP), to accurately identify eccDNA from long-read Nanopore sequencing data. Compared to other pipelines, ECCFP significantly improves detection sensitivity, accuracy, and runtime efficiency. The process includes raw data quality control, trimming of adapters and barcodes, alignment to a reference genome, and identification of eccDNA, with detailed results encompassing accurate positioning of eccDNA, consensus sequences, and variants of individual eccDNA.

Cellulose synthase complexes (CSCs) play a central role in plant cell wall formation. Their dynamic behavior at the plasma membrane leads to the deposition of cellulose microfibrils into the apoplastic space, thereby shaping the architecture and mechanical properties of the cell wall. Although previous imaging studies have provided important insights into CSC dynamics and localization, standardized and reproducible workflows for quantitative measurements of CSC speed and density remain limited. Here, we present a reproducible live-cell imaging and analysis workflow for quantifying the speed and density of fluorescently labeled CSCs at the plasma membrane in Arabidopsis thaliana. The protocol integrates optimized spinning-disk confocal imaging, surface-based projection of z-stack recordings, automated detection of diffraction-limited CSCs foci, and kymograph-based speed measurements using freely available tools in Fiji. While selected steps, such as region of interest definition and parameter selection for spot detection or trajectory analysis, remain user-guided, these decisions are constrained to well-defined stages within an otherwise standardized pipeline, thereby reducing variability and improving reproducibility across experiments. The workflow has been validated across multiple tissues, reporter lines, genetic backgrounds, and perturbation conditions in Arabidopsis and enables robust comparative analysis of CSC dynamics. Beyond CSCs, this workflow is expected to be adaptable to other fluorescently labeled proteins that appear as diffraction-limited foci at or near the plasma membrane.

Organelle abundance is a key microscopic readout of organelle formation and, in many cases, function. Quantification of organelle abundance using confocal microscopy requires estimating their area based on the fluorescence intensity of compartment-specific markers. This analysis usually depends on a user-defined intensity threshold to distinguish organelle regions from the surrounding cytoplasm, which introduces potential bias and variability. To address this issue, we present a machine learning–assisted algorithm that allows for the quantification of organelle density using the open-source Fiji platform and WEKA segmentation. Our method enables the automated quantification of organelle number, area, and density by learning from training data. This standardizes threshold selection and minimizes user intervention. We demonstrate the utility of this approach for both membrane and non-membrane organelles, such as peroxisomes, lipid droplets, and stress granules, in human cells and whole fish samples.

Spatial proteomics enables the mapping of protein distribution within tissues, which is crucial for understanding cellular functions in their native context. While spatial transcriptomics has seen rapid advancement, spatial proteomics faces challenges due to protein non-amplifiability and mass spectrometry sensitivity limitations. This protocol describes a sparse sampling strategy for spatial proteomics (S4P) that combines multi-angle tissue strip microdissection with deep learning–based image reconstruction. The method achieves whole-tissue slice coverage with significantly reduced sampling requirements, enabling mapping of over 9,000 proteins in mouse brain tissue at 525 μm resolution within 200 h of mass spectrometry time. Key advantages include reduced sample processing time, deep proteome coverage, and applicability to centimeter-sized tissue samples.

The morphology of single-neuron axonal projections is critical for deciphering neural circuitry and information flow in the brain. Yet, manually reconstructing these complex, long-range projections from high-throughput whole-brain imaging data remains an exceptionally labor-intensive and time-consuming task. Here, we developed a points assignment-based method for axonal reconstruction, named PointTree. PointTree enables the precise identification of the individual axons from densely packed axonal population using a minimal information flow tree model to suppress the snowball effect of reconstruction errors. In this protocol, we have elaborated on how to configure the required environment for PointTree software, prepare suitable data for it, and run the software. This protocol can assist neuroscience researchers in more easily and rapidly obtaining the reconstruction results of neuronal axons.

Evaluating single-domain antibody cooperativity is essential for developing potent, escape-resistant antiviral biologics. Here, we present a protocol that reproducibly quantifies functional synergy between neutralizing nanobody pairs in standardized viral infectivity assays. Controlled automated liquid handling prepares two-dimensional concentration matrices, minimizing pipetting variance and systematic error. Neutralization data are fitted using quantitative models that independently estimate potency, cooperativity, and efficacy to distinguish additive, synergistic, and antagonistic effects between nanobody pairs. Replicated measurements enable statistically interpretable parameter estimates, supporting robust evaluation of combinatorial nanobody therapeutics with commonly available equipment and open-source analysis tools. This framework is broadly applicable to assessing cooperative effects among other classes of binding or inhibitory molecules, facilitating systematic discovery of synergistic combinations.

RNA-binding protein (RBP)–RNA interactions are fundamental for gene regulation and cellular homeostasis. Ataxin-2 is an RBP that has been shown to play an instrumental role in pathophysiological processes by binding to mRNA. Methods such as RNA immunoprecipitation (RIP), cross-linking immunoprecipitation (CLIP), and their variants can be used to study the interactions between Ataxin-2 and its targets, although their high sample requirements and labor-intensive workflows can limit their widespread use. RNA editing-based approaches, such as targets of RBPs identified by editing (TRIBE), provide effective alternatives. TRIBE enables transcriptome-wide identification of RBP targets by inducing site-specific adenosine-to-inosine (A-to-I) editing, which is subsequently detected through high-throughput RNA sequencing in both in vivo and in vitro systems. Compared to in vivo models, cell lines offer a rapid and flexible experimental design. Drosophila S2 cells are a commonly used insect cell line to investigate RNA–protein dynamics and serve as a versatile platform for studying RBP function. Here, we describe a protocol used for identifying RNA targets of Ataxin-2, a versatile RBP involved in post-transcriptional and translational regulation, in S2 cells using TRIBE. This method allows rapid, efficient, and reliable identification of Ataxin-2-associated RNA targets and can be readily applied to other RBPs.