Biology subjects

Bai, Y.

Publications and source records attributed to Bai, Y..

At least 127 records · Page 7Linked to original sources

bioRxiv2023

Skeletal Muscle Satellite Cells Co-Opt the Tenogenic Gene Scleraxis to Instruct Regeneration

Skeletal muscles connect bones and tendons for locomotion and posture. Understanding the regenerative processes of muscle, bone and tendon is of importance to basic research and clinical applications. Despite their interconnections, distinct transcription factors have been reported to orchestrate each tissues developmental and regenerative processes. Here we show that Scx expression is not detectable in adult muscle stem cells (also known as satellite cells, SCs) during quiescence. Scx expression begins in activated SCs and continues throughout regenerative myogenesis after injury. By SC-specific Scx gene inactivation (ScxcKO), we show that Scx function is required for SC expansion/renewal and robust new myofiber formation after injury. We combined single-cell RNA-sequencing and CUT&RUN to identify direct Scx target genes during muscle regeneration. These target genes help explain the muscle regeneration defects of ScxcKO, and are not overlapping with Scx-target genes identified in tendon development. Together with a recent finding of a subpopulation of Scx-expressing connective tissue fibroblasts with myogenic potential during early embryogenesis, we propose that regenerative and developmental myogenesis co-opt the Scx gene via different mechanisms.

cell biology↗

bioRxiv2023

Spatially resolved molecular and cellular atlas of the mouse brain

A comprehensive atlas of genes, cell types, and their spatial distribution across a whole mammalian brain is fundamental for understanding function of the brain. Here, using snRNA-seq and Stereo-seq techniques, we generated a mouse brain atlas with spatial information for 308 cell clusters with single-cell resolution involving over 6 million cells as well as for 29,655 genes. We have identified new astrocyte clusters, and demonstrated that distinct cell clusters exhibit preference for cortical subregions. In addition, we identified 155 genes exhibiting regional specificity in the brainstem, and 513 long non-coding RNA exhibited regional specificity in the adult brain. Parcellation of brain regions based on spatial transcriptomic information showed large overlap with that by traditional method. Furthermore, we have uncovered 411 transcription factor regulons with spatiotemporal specificity during development. Thus, our study has discovered genes and regulon with spatiotemporal specificity, and provided a high-resolution spatial transcriptomic atlas of the mouse brain.

neuroscience↗

bioRxiv2023

iTraNet: A Web-Based Platform for integrated Trans-Omics Network Visualization and Analysis

ABSTRACTA major goal in biology is to comprehensively understand molecular interactions within living systems. Visualization and analysis of biological networks play crucial roles in understanding these biochemical processes. Biological networks include diverse types, from gene regulatory networks and protein-protein interactions (PPIs) to metabolic networks. Metabolic networks include substrates, products, and enzymes, which are regulated by allosteric mechanisms and gene expression. Given this complexity, there is a pressing need to investigate trans-omics networks that include these various regulations to understand living systems. However, analyzing various omics layers is laborious due to the diversity of databases and the intricate nature of network analysis. We developed iTraNet, a user-friendly interactive web application that visualizes and analyzes trans-omics networks involving four major types of networks: gene regulatory networks (including transcription factor, microRNA, and mRNA); PPIs; metabolic networks (including enzyme, mRNA, and metabolite); and metabolite exchange networks (including transporter, mRNA, and metabolite). Using iTraNet, we found that in wild-type mice, hub molecules within the network tended to respond to glucose administration, whereas in ob/ob mice, this tendency disappeared. With its ability to facilitate network visualization and analysis, we anticipate that iTraNet will help researchers gain insights into biological systems. iTraNet is available at (https://transomics.streamlit.app/). GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=75 SRC="FIGDIR/small/569499v1_ufig1.gif" ALT="Figure 1"> View larger version (21K): org.highwire.dtl.DTLVardef@12bb1c1org.highwire.dtl.DTLVardef@1a326b7org.highwire.dtl.DTLVardef@42cfc9org.highwire.dtl.DTLVardef@57aee6_HPS_FORMAT_FIGEXP M_FIG C_FIG

systems biology↗

bioRxiv2023

Diagnostic images for mild cognitive impairment reveal biomarker status and abnormal scene processing

Research on the impairment of episodic memory in Alzheimers disease often focuses on the processes of memory rather than the content of the specific images being remembered. We recently showed that patients with mild cognitive impairment (MCI), Stage 3 of Alzheimers disease, can memorize certain images quite well, suggesting that episodic memory is not uniformly impaired. Certain images, on the other hand, could not be memorized by MCI patients and were instead diagnostic for distinguishing MCI from healthy older adults. In this study, we investigate whether poor memory for diagnostic images is related to impaired neural processing in specific brain regions due to Alzheimers biomarker pathology. 64 healthy controls and 48 MCI participants in the DELCODE dataset performed a visual scene memory task during fMRI, with CSF Alzheimers disease biomarker data collected (i.e., amyloid and tau biomarkers). We found that diagnostic images have larger behavior-biomarker correlations for total tau, phospho-tau, A{beta}42/A{beta}40, A{beta}42/phospho-tau compared to non-diagnostic images, suggesting that memory for these specific images are more affected by Alzheimers disease pathology. The fMRI data revealed an interaction effect between group membership (healthy control / MCI) and image diagnosticity (diagnostic / non-diagnostic scene images), with MCI participants having higher activation in scene processing regions (parahippocampal place area, retrosplenial cortex and occipital place area) for diagnostic images than non-diagnostic images. In contrast, healthy controls showed no differences in processing between diagnostic and non-diagnostic images. These results suggest that MCI individuals may engage in inefficiently heightened encoding activation for these diagnostic images. Our results show that special "diagnostic" images exist that can reveal amyloid and tau pathology and differences in neural activity in scene regions.

neuroscience↗

bioRxiv2023

Discovering Novel Prognostic Biomarkers of Hepatocellular Carcinoma using eXplainable Artificial Intelligence

Hepatocellular carcinoma (HCC) remains a global health challenge with high mortality rates, largely due to late diagnosis and suboptimal efficacy of current therapies. With the imperative need for more reliable, non-invasive diagnostic tools and novel therapeutic strategies, this study focuses on the discovery and application of novel genetic biomarkers for HCC using explainable artificial intelligence (XAI). Despite advances in HCC research, current biomarkers like Alpha-fetoprotein (AFP) exhibit limitations in sensitivity and specificity, necessitating a shift towards more precise and reliable markers. This paper presents an innovative XAI framework to identify and validate key genetic biomarkers for HCC prognosis. Our methodology involved analyzing clinical and gene expression data to identify potential biomarkers with prognostic significance. The study utilized robust AI models validated against extensive gene expression datasets, demonstrating not only the predictive accuracy but also the clinical relevance of the identified biomarkers through explainable metrics. The findings highlight the importance of biomarkers such as TOP3B, SSBP3, and COX7A2L, which were consistently influential across multiple models, suggesting their role in improving the predictive accuracy for HCC prognosis beyond AFP. Notably, the study also emphasizes the relevance of these biomarkers to the Hispanic population, aligning with the larger goal of demographic-specific research. The application of XAI in biomarker discovery represents a significant advancement in HCC research, offering a more nuanced understanding of the disease and laying the groundwork for improved diagnostic and therapeutic strategies.

cancer biology↗

bioRxiv2023

Trans-omic analysis reveals opposite metabolic dysregulation between feeding and fasting in liver associated with obesity

Dysregulation of liver metabolism associated with obesity during feeding and fasting leads to the breakdown of metabolic homeostasis. However, the underlying mechanism remains unknown. Here, we measured multi-omics data in the liver of wild-type and leptin-deficient obese (ob/ob) mice at ad libitum feeding, and constructed a differential regulatory trans-omic network of metabolic reactions. We compared the trans-omic network at feeding with that at 16 h-fasting constructed in our previous study. Intermediate metabolites in glycolytic and nucleotide metabolism decreased in ob/ob mice at feeding but increased at fasting. Allosteric regulation reversely shifted between feeding and fasting, generally showing activation at feeding while inhibition at fasting in ob/ob mice. Transcriptional regulation was similar between feeding and fasting, generally showing inhibiting transcription factor regulations, activating enzyme protein regulations in ob/ob mice. The opposite metabolic dysregulation between feeding and fasting characterizes breakdown of metabolic homeostasis associated with obesity.

systems biology↗

bioRxiv2023

Image guided construction of a common coordinate framework for spatial transcriptome data

Spatial transcriptomics is a powerful technology for high-resolution mapping of gene expression in tissue samples, enabling a molecular level understanding of tissue architecture. The acquisition entails dissecting and profiling micron-thick tissue slices, with multiple slices often needed for a comprehensive study. However, the lack of a common coordinate framework (CCF) among slices, due to slicing and displacement variations, can hinder data analysis, making data comparison and integration challenging, and potentially compromising analysis accuracy. Here we present a deep learning algorithm STaCker that unifies the coordinates of transcriptomic slices via an image registration process. STaCker derives a composite image representation by integrating tissue image and gene expressions that are transformed to be resilient to noise and batch effects. Trained exclusively on diverse synthetic data, STaCker overcomes the training data scarcity and is applicable to any tissue type. Its performance on various benchmarking datasets shows a significant increase in spatial concordance in aligned slices, surpassing existing methods. STaCker also successfully harmonizes multiple real spatial transcriptome datasets. These results indicate that STaCker is a valuable computational tool for constructing a CCF with spatial transcriptome data.

bioinformatics↗

bioRxiv2023

The human LIN28B nucleosome is inherently pre-positioned for efficient binding of multiple OCT4s without H3 K27 acetylation

Pioneer transcription factors possess the unique ability to access DNA within tightly packed chromatin structures, playing pivotal roles in cell differentiation and reprogramming. However, their precise mechanism for recognizing nucleosomes has remained mystery. Recent structural and biochemical investigations into the binding interactions between the human pioneer factor OCT4 and the LIN28B nucleosome by Sinha et al.1 and Guan et al.2 have yielded conflicting results regarding nucleosome positioning, nucleosomal DNA unwrapping, binding cooperativity, and the role of N-terminal tail of OCT4. In this study, we undertook a comparative analysis of these two research efforts and delved into the factors contributing to the observed discrepancies. Our investigation unveiled that the utilization of human and Xenopus laevis core histones, along with a discrete two-step salt dialysis method, led to distinct positioning of DNA within reconstituted LIN28B nucleosomes. Additionally, our reanalysis of the electrophoretic mobility shift assay data showed that H3 K27 acetylation did not increase OCT4 binding to the internal sites of the nucleosome when normalized to input; instead, it promoted sample aggregation. Thus, the available experimental data support the notion that the human LIN28B nucleosome is pre-positioned for efficient binding with multiple OCT4s, and there is no compelling evidence for its regulation by histone modifications.

biochemistry↗

bioRxiv2023

Injury induced renal fibrosis promotes cystogenesis and cyst growth in adult mice with autosomal dominant polycystic kidney disease

Autosomal dominant polycystic kidney disease (ADPKD) is a life-long disease and caused by mutations in PKD1 or PKD2 gene. Fibrosis is a hallmark of chronic kidney disease and is positively correlated with renal cyst growth, however the role of fibrosis in ADPKD is still controversial. In this study, we established renal fibrosis by toxic or surgical injuries in adult mice, and Pkd gene was inactivated at different time point before or after renal injury according to the pattern of fibrosis progression in different injury models. Here we showed that renal injury before or after Pkd gene inactivation can both induce renal cysts in adult Pkd1 or Pkd2 mice, and the extent of cystic burden was tightly correlated with the baseline levels of fibrosis when three hits (injury and gene inactivation) occurred. Inactivation of Pkd1 gene at the recovery stage after surgery induced less renal cysts in adult Pkd1 mice. Enhanced renal fibrosis by repeated toxic injuries before gene inactivation accelerated renal cyst growth in Pkd1 mice. We further showed that the rate of cyst formation at the early stage in adult Pkd1 mice was positively correlated with the baseline levels of renal fibrosis. Finally, we showed that conditional knockout of Ezh2 gene attenuated renal fibrosis and cyst growth in adult Pkd1 mice with pre-existing renal fibrosis. We conclude that the fibrotic response after renal injury is a driving force for renal cyst formation and growth in adult kidneys and inhibition of renal fibrosis through targeting EZH2 might be new therapeutic strategy for adult ADPKD. Importantly, our study suggests that there is a time window for intervention upon acute kidney injury in adult ADPKD patients. Translational StatementAutosomal dominant polycystic kidney disease (ADPKD) is a life-long disease and caused by mutations in PKD1 or PKD2 gene. Fibrosis is a hallmark of chronic kidney disease and is positively correlated with renal cyst growth, however the role of renal fibrosis in ADPKD is controversial. In this study, we found that renal cysts were formed in adult Pkd1 or Pkd2 mice with established renal fibrosis induced by toxic or ischemia reperfusion injuries. Cyst formation or growth in adult ADPKD mice was tightly correlated with baseline levels of renal fibrosis after third hits. Enhanced renal fibrosis before Pkd1 gene deletion in adult mice accelerated cyst growth. Inhibition of renal fibrosis through targeting EZH2 delayed cyst growth in adult ADPKD mice. Thus, renal fibrosis is a trigger of cyst formation and growth in adult ADPKD mice, and therapeutically targeting EZH2 might be new strategy to treat adult patients with ADPKD. Our study suggests that there is a time window for intervention upon acute kidney injury in adult ADPKD patients.

pathology↗

bioRxiv2023

Single-cell bisulfite-free 5mC and 5hmC sequencing with high sensitivity and scalability

Existing single cell bisulfite-based DNA methylation analysis is limited by low DNA recovery, and the measurement of 5hmC at single-base resolution remains challenging. Here we present a bisulfite-free single-cell whole-genome 5mC and 5hmC profiling technique, named Cabernet, which can characterize 5mC and 5hmC at single-base resolution with high genomic coverage. Cabernet utilizes Tn5 transposome for DNA fragmentation, which enables the discrimination between different alleles for measuring hemi-methylation status. Using Cabernet, we revealed the 5mC, hemi-5mC and 5hmC dynamics during early mouse embryo development, uncovering genomic regions exclusively governed by active or passive demethylation. We show that hemi-methylation status can be used to distinguish between pre- and post-replication cells, enabling more efficient cell grouping when integrated with 5mC profile. The property of Tn5 naturally enables Cabernet to achieve high-throughput single-cell methylome profiling, where we probed mouse cortical neurons and embryonic day 7.5 (E7.5) embryos, and constructed the library for thousands of single cells at high efficiency, demonstrating its potential for analyzing complex tissues at substantially low cost. Together, we present a new way of high-throughput methylome and hydroxymethylome detection at single-cell resolution, enabling efficient analysis of epigenetic status of biological systems with complicated nature such as neurons and cancer cells. Significance StatementMost of current methylation profiling techniques rely on bisulfite treatment, which suffers low DNA recovery. The technique proposed in this study, named Cabernet, can be used to measure 5mC and 5hmC at single-base resolution with high genomic coverage. By using Tn5 transposome, hemi-methylation status can be measured and high-throughput methylome profiling can be achieved. Together, it provides an efficient way to analyze the epigenetic landscape of complicated biological systems.

genomics↗

bioRxiv2023

SARS-CoV-2 N protein-induced Dicer, XPO5, SRSF3, and hnRNPA3 downregulation causes pneumonia

Age is a major risk factor for coronavirus disease (COVID-19)-associated severe pneumonia and mortality; however, the underlying mechanism remains unclear. Herein, we investigated whether age-related deregulation of RNAi components and RNA splicing factors affects COVID-19 severity. Decreased expression of RNAi components (Dicer and XPO5) and splicing factors (SRSF3 and hnRNPA3) correlated with increased severity of COVID-19 and SARS-CoV-2 nucleocapsid (N) protein-induced pneumonia. N protein induced autophagic degradation of Dicer, XPO5, SRSF3, and hnRNPA3, repressing miRNA biogenesis and RNA splicing and inducing DNA damage, proteotoxic stress, and pneumonia. Dicer, XPO5, SRSF3, and hnRNPA3 were downregulated with age in mouse lung tissues. Older mice experienced more severe N protein-induced pneumonia than younger mice. However, treatment with a poly(ADP-ribose) polymerase inhibitor (PJ34) or aromatase inhibitor (anastrozole) relieved N protein-induced pneumonia by restoring Dicer, XPO5, SRSF3, and hnRNPA3 expression. These findings will aid in developing improved treatments for SARS-CoV-2-associated pneumonia.

microbiology↗

bioRxiv2023

Age-related STAT3 signaling regulates severity of respiratory syncytial viral infection in human bronchial epithelial cells

Respiratory syncytial virus (RSV) can cause severe disease especially in infants; however, mechanisms of age-associated disease severity remain elusive. Here, employing human bronchial epithelium models generated from tracheal aspirate-derived basal stem cells of neonates and adults, we investigated whether age regulates RSV-epithelium interaction to determine disease severity. We show that following RSV infection, only neonatal epithelium model exhibited cytopathy and mucus hyperplasia, and neonatal epithelium had more robust viral spread and inflammatory responses than adult epithelium. Mechanistically, RSV-infected neonatal ciliated cells displayed age-related impairment of STAT3 activation, rendering susceptibility to apoptosis, which facilitated viral spread. In contrast, SARS-CoV-2 infection of ciliated cells had no effect on STAT3 activation and was not affected by age. Taken together, our findings identify an age-related and RSV-specific interaction with neonatal bronchial epithelium that critically contributes to severity of infection, and STAT3 activation offers a potential strategy to battle severe RSV disease in infants.

cell biology↗

bioRxiv2023

A Small Tau Fragment Specifically Templates Four Repeat Tau Aggregates Through Multiple Generations

Prion-like spread of disease-specific tau conformers is a hallmark of all tauopathies. A 19-residue probe peptide containing a P301L mutation and spanning the R2/R3 splice junction of tau, folds and stacks into seeding-competent fibrils and induces aggregation of 4R, but not 3R tau. These tau peptide fibrils propagate aggregated intracellular tau over multiple generations, have a high {beta}- sheet content, a colocalized lipid signal, and adopt a well-defined U-shaped fold found in 4R tauopathy brain-derived fibrils. Fully atomistic replica exchange molecular dynamics (MD) simulations were used to compute the free energy landscapes of the conformational ensemble of the peptide monomers. These identified an aggregation-prohibiting {beta}-hairpin structure and an aggregation-competent U-fold unique to 4R tauopathy fibrils. Guided by MD simulations, we identified that the N-terminal-flanking residues to PHF6, which slightly vary between 4R and 3R isoforms, modulate seeding. Strikingly, when a single amino acid switch at position 305 replaced the serine of 4R tau with a lysine from the corresponding position in the first repeat of 3R tau, the seeding induced by the 19-residue peptide was markedly reduced. Conversely, a 4R tau mimic with three repeats, prepared by replacing those amino acids in the first repeat with those amino acids uniquely present in the second repeat, recovered aggregation when exposed to the 19- residue peptide. These peptide fibrils function as partial prions to recruit naive 4R tau--ten times the length of the peptide--and serve as a critical template for 4R tauopathy propagation. These results hint at opportunities for tau isoform-specific therapeutic interventions. Significance StatementA structural motif corresponding to a short junction sequence spanning R2 and R3 forms fibrils that adopt a fold characteristic of 4R tauopathy fibrils and induces misfolding of the larger tau protein with loss of microtubule binding and a prion-like specificity for 4R tau. Simulations, validated experimentally, pinpointed the specific amino acids in the peptide that can toggle its properties between aggregation competent and incompetent. The modifications suggest design principles for a therapeutic intervention potentially capable of disaggregating tau or preventing its aggregation in the 4R tauopathies.

molecular biology↗

bioRxiv2023

Sequential breakdown of the complex Cf-9 leaf mould resistance locus in tomato by Fulvia fulva

O_LILeaf mould, caused by Fulvia fulva, is a devastating disease of tomato plants. In many commercial tomato cultivars, resistance to this disease is governed by the Cf-9 locus, which comprises five paralogous genes (Cf-9A-9E) that encode receptor-like proteins. Two of these proteins contribute to resistance: Cf-9C recognizes the previously identified F. fulva effector Avr9 and provides resistance during all plant growth stages, while Cf-9B recognises the yet-unidentified F. fulva effector Avr9B and provides mature plant resistance only. In recent years, F. fulva strains have emerged that have overcome the Cf-9 locus, with Cf-9C circumvented through Avr9 deletion. To understand how Cf-9B is circumvented, we set out to identify Avr9B. C_LIO_LIComparative genomics, in planta transient expression assays and gene complementation experiments were used to identify Avr9B, while gene sequencing was used to assess Avr9B allelic variation across a worldwide strain collection. C_LIO_LIA strict correlation between Avr9 deletion and resistance-breaking mutations in Avr9B was observed in strains recently collected from Cf-9 cultivars, whereas Avr9 deletion but no mutations in Avr9B were observed in older strains. C_LIO_LIThis research showcases how F. fulva has evolved to sequentially break down the two functional resistance genes of the complex Cf-9 locus and highlights that this locus now has limited value for controlling leaf mould disease in worldwide commercial tomato production. C_LI

microbiology↗

bioRxiv2023

Structural mechanism of synergistic targeting of the CX3CR1 nucleosome by PU.1 and C/EBP

Pioneer transcription factors are vital for cell fate changes. PU.1 and C/EBP work together to regulate hematopoietic stem cell differentiation. However, how they recognize in vivo nucleosomal DNA targets remain elusive. Here we report the structures of the nucleosome containing the mouse genomic CX3CR1 enhancer DNA and its complexes with PU.1 alone and with both PU.1 and the C/EBP DNA binding domain. Our structures reveal that PU.1 binds the DNA motif at the exit linker, shifting 17 bp of DNA into the core region through interactions with H2A, unwrapping [~]20 bp of nucleosomal DNA. C/EBP binding, aided by PU.1s repositioning, unwraps [~]25 bp entry DNA. The PU.1 Q218H mutation, linked to acute myeloid leukemia, disrupts PU.1-H2A interactions. PU.1 and C/EBP jointly displace linker histone H1 and open the H1-condensed nucleosome array. Our study unveils how two pioneer factors can work cooperatively to open closed chromatin by altering DNA positioning in the nucleosome.

biophysics↗

bioRxiv2023

Chromosome level assembly of wild spinach provides insights into the divergence of homo- and heteromorphic plant sex-chromosomes.

BackgroundCultivated spinach (Spinacia oleracea) is a highly nutritional crop species of great economical value that belongs to a genus of dioecious plant species with both homomorphic and heteromorphic sex chromosomes. The wild spinach species Spinacia turkestanica and Spinacia tetrandra are important genetic sources for improving cultivated spinach and excellent material for studying sex chromosome evolution in plants. However, until now there were no publicly available genome assemblies for these species. ResultsHere we sequenced and assembled the genomes of S. turkestanica and S. tetrandra and performed a tri-way comparative analysis with S. oleracea. We show that many abiotic- and biotic stress related gene clusters have expanded through tandem duplication in S. tetrandra after it diverged from the S. turkestanica - S. oleracea lineage. Focussing on the sex chromosomes we found that the previously identified inversion distinguishing the S. oleracea male- and female-SEX DETERMINING REGIONs (SDRs) is conserved in S. turkestanica. Although, the SDRs of these two species coincides with the PSEUDO AUTOSOMAL REGION of S. tetrandra the gene content is only partially conserved and the genetic factors determining sex in these species might differ. Finally, we show that recombination suppression between the S. tetrandra X- and Y-chromosomes resulted in a highly degenerated Y-chromosome and started before the species diverged from S. turkestanica and S. oleracea. ConclusionsWe expect that the novel wild spinach species genomes are of great value to the breeding community and evolutionary biologist especially focussing on the evolution of sex chromosomes in plants.

evolutionary biology↗

bioRxiv2023

Cross-regulation between proteome reallocation and metabolic flux redistribution governs bacterial growth transition kinetics

Bacteria need to adjust their metabolism and protein synthesis simultaneously to adapt to changing nutrient conditions. Its still a grand challenge to predict how cells coordinate such adaptation due to the cross-regulation between the metabolic fluxes and the protein synthesis. Here we developed a dynamic Constrained Allocation Flux Balance Analysis method (dCAFBA), which integrates flux-controlled proteome allocation and protein limited flux balance analysis. This framework can predict the redistribution dynamics of metabolic fluxes without requiring detailed enzyme parameters. We reveal that during nutrient up-shifts, the calculated metabolic fluxes change in agreement with experimental measurements of enzyme protein dynamics. During nutrient down-shifts, we uncover a switch of metabolic bottleneck from carbon uptake proteins to metabolic enzymes, which disrupts the coordination between metabolic flux and their enzyme abundance. Our method provides a quantitative framework to investigate cellular metabolism under varying environments and reveals insights into bacterial adaptation strategies.

systems biology↗

bioRxiv2023

MAPS: Pathologist-level cell type annotation from tissue images through machine learning

Highly multiplexed protein imaging is emerging as a potent technique for analyzing protein distribution within cells and tissues in their native context. However, existing cell annotation methods utilizing high-plex spatial proteomics data are resource intensive and necessitate iterative expert input, thereby constraining their scalability and practicality for extensive datasets. We introduce MAPS (Machine learning for Analysis of Proteomics in Spatial biology), a machine learning approach facilitating rapid and precise cell type identification with human-level accuracy from spatial proteomics data. Validated on multiple in-house and publicly available MIBI and CODEX datasets, MAPS outperforms current annotation techniques in terms of speed and accuracy, achieving pathologist-level precision even for challenging cell types, including tumor cells of immune origin. By democratizing rapidly deployable and scalable machine learning annotation, MAPS holds significant potential to expedite advances in tissue biology and disease comprehension.

cancer biology↗