Biology subjects

Tan, C.

Publications and source records attributed to Tan, C..

At least 55 records · Page 3Linked to original sources

bioRxiv2024

FoldToken3: Fold Structures Worth 256 Words or Less

Protein structure tokenization has attracted increasing attention in both protein representation learning and generation. While recent work, like FoldToken2 and ESM3, has achieved good reconstruction performance, the compressoin ratio is still limited. In this work, we propose FoldToken3, a novel protein structure tokenization method that can compress protein structures into 256 tokens or less and ensure the reconstruction quality comparable to FoldToken2. To the best of our knowledge, FoldToken3 is the most efficient, light-weight, and compression-friendly protein structure tokenization method. And it will benifit a wide range of protein structure-related tasks, such as protein structure alignment, generation, and representation learning. The work is still in progress and the code will be available upon acceptance.

molecular biology↗

bioRxiv2024

High efficient chromatin conformation capture without pre-enrichment (HiChew) in single cells

This study presents HiChew, a cutting-edge technique for high-efficiency chromatin conformation capture in single cells, without the need for pre-enrichment. This unique approach minimizes the risk of cell or DNA loss. When compared to Dip-C, HiChew captures valid pairs with 4-8 times more efficiency, reducing wastage and saving significant sequencing budget. Furthermore, HiChew delivers a lower false positive ratio, ensuring data accuracy. It also achieves more contacts per cell, enhancing resolution in single cell. HiChews superior performance not only enhances single-cell Hi-C but also streamlines conventional Hi-C, making it more robust than conventional HiC methods. This study also unveils a fascinating mechanism of gene activation in the B compartment of chromatin, providing insight into the elusive aspect of gene expression within this region.

genomics↗

bioRxiv2024

Exploring DNA movement through the application of droplet based high efficient chromatin conformation capture (DropHiChew) and loop velocity

This study presents a novel approach to understanding DNA movement dynamics through the development of a droplet-based, high-efficient chromatin conformation capture method, known as DropHiChew, and a new algorithm, loop velocity. DropHiChew, a user-friendly and cost-effective technique, employs the 10X single-cell systems allowing for easy experimental implementation. The loop velocity algorithm, on the other hand, enables the estimation of the speed and direction of cell development, providing a dynamic perspective on chromatin movement. Even with shallow sequencing, our loop velocity algorithm accurately gauges the trajectory of DNA motion and cellular states. The combined use of DropHiChew and loop velocity offers potential for a wide array of applications in future chromatin capture studies, including disease modeling, cellular differentiation studies, and developmental biology.

genomics↗

bioRxiv2024

Heterogenous organization in condensates of multiple transcription factors in embryonic stem cells

Biomolecular condensates formed via liquid-liquid phase separation are ubiquitous in cells, especially in the nucleus. While condensates containing one or two kinds of biomolecules have been relatively well characterized, those with more heterogenous biomolecular components and interactions between biomolecules inside are largely unknown. This study used residue-resolution molecular dynamics simulations to investigate heterogeneous protein assemblies that include four master transcription factors in mammalian embryonic stem cells: Oct4, Sox2, Klf4, and Nanog. Simulations of the mixture systems showed highly heterogeneous and dynamic behaviors; the condensates mainly contained Sox2, Klf4, and Nanog, while Oct4 was dissolved into the dilute phase. Condensates consisted of loosely interacting clusters in which Klf4 was the most abundant in the cores. We suggest that Klf4 serves as a scaffold of the condensate where Sox2 and Nanog are bound to stabilize the condensate, whereas Oct4 is moderately recruited to the condensate, serving as a client mainly via its interaction with Sox2. Biological significanceIn eukaryotes, transcription is known to be regulated by many regulatory factors such as transcription factors, co-activators, and RNA polymerases, but precise molecular mechanisms of regulation remain obscured. A recently proposed model suggests that transcription-related proteins condense via liquid-liquid phase separation (LLPS) using their intrinsically disordered regions, which serves to control transcription. Master transcription factors in mammalian embryonic stem cells have been a model system. It was revealed that several master transcription factors exhibit LLPS by themselves, but dynamics and molecular structure of these proteins in their mixture have not been well addressed. In this study, we study molecular structures of condensates in a mixture of four master transcription factors, Oct4, Sox2, Klf4, and Nanog, via molecular dynamics simulation. We found that the three transcription factors Sox2, Klf4, and Nanog form mixed condensates, while Oct4 was largely dissolved. Klf4 mainly served as a scaffold of the condensate. The three proteins formed micelle-like structures as was recently found in the Nanog condensate. The condensates weakly recruited Oct4. Formation of heterogeneous condensates may provide fertile local environments in cells.

biophysics↗

bioRxiv2024

FoldToken2: Learning compact, invariant and generative protein structure language

The equivariant nature of 3D coordinates has posed long term challenges in protein structure representation learning, alignment, and generation. Can we create a compact and invariant language that equivalently represents protein structures? Towards this goal, we propose FoldToken2 to transfer equivariant structures into discrete tokens, while maintaining the recoverability of the original structures. From FoldToken1 to FoldToken2, we improve three key components: (1) invariant structure encoder, (2) vector-quantized compressor, and (3) equivariant structure decoder. We evaluate FoldToken2 on the protein structure reconstruction task and show that it outperforms previous FoldToken1 by 20% in TMScore and 81% in RMSD. FoldToken2 is likely the first method that works well for both single-chain and multi-chain protein structure quantization. We believe that FoldToken2 will inspire further improvement in protein structure representation, alignment, and generation tasks. Online example is available at Colab.

molecular biology↗

bioRxiv2024

The phylogenetic position of the Yunxian cranium elucidates the origin of Dragon Man and the Denisovans

Diverse Middle Pleistocene forms of Homo coexisted in Africa, Europe, and Asia, and its controversial whether these fossil humans represent different species or clades. The [~]1 Ma old Yunxian 2 fossil from China is crucial for understanding the cladogenesis of Homo and the origin of Homo sapiens. Here, we restored and reconstructed the distorted Yunxian 2 cranium using new technology. The results show that this cranium displays mosaic features of plesiomorphy and apomorphy. Geometric-morphometric, phylogenetic and Bayesian tip-dating analyses including the reconstructed Yunxian 2 suggest that it is an early member of the Asian Homo longi clade, which probably includes the Denisovans, and is the sister group to the Homo sapiens clade. Both the H. sapiens and H. longi clades had deep roots extending beyond the Middle Pleistocene, and the basal position of the Yunxian fossil cranium suggests that it represents a population lying close to the last common ancestor of the two clades. One-Sentence SummaryThe newly-reconstructed Yunxian 2 cranium represents a basal member of the Homo longi and Denisovan clade, and probably lies close to the last common ancestor of that clade and the clade of H. sapiens.

evolutionary biology↗

bioRxiv2024

High Frequency Terahertz Stimulation Alleviates Neuropathic Pain by Inhibiting the Pyramidal Neuron Activity in the Anterior Cingulate Cortex of mice

Neuropathic pain (NP) is caused by a lesion or disease of the somatosensory system and is characterized by abnormal hypersensitivity to stimuli and nociceptive responses to non-noxious stimuli, affecting approximately 7-10% of the general population. However, current first-line drugs like non-steroidal anti-inflammatory agents and opioids have limitations, including dose-limiting side effects, dependence, and tolerability issues. Therefore, developing new interventions for the management of NP is urgent. In this study, we discovered that the high-frequency terahertz stimulation (HFTS) at approximate 36 THz effectively alleviates NP symptoms in mice with spared nerve injury. Computational simulation suggests that the frequency resonates with the carbonyl group in the filter region of Kv1.2 channels, facilitating the translocation of potassium ions. In vivo and in vitro results demonstrate that HFTS reduces the excitability of pyramidal neurons in the anterior cingulate cortex through enhancing the voltage-gated K+ and also the leak K+ conductance. This research presents a novel optical intervention strategy with terahertz waves for the treatment of NP and holds promising application in other nervous system diseases.

neuroscience↗

bioRxiv2024

Deciphering Endothelial and Mesenchymal Organ Specification in Vascularized Lung and Intestinal Organoids

To investigate the co-development of vasculature, mesenchyme, and epithelium crucial for organogenesis and the acquisition of organ-specific characteristics, we constructed a human pluripotent stem cell-derived organoid system comprising lung or intestinal epithelium surrounded by organotypic mesenchyme and vasculature. We demonstrated the pivotal role of co-differentiating mesoderm and endoderm via precise BMP regulation in generating multilineage organoids and gut tube patterning. Single-cell RNA-seq analysis revealed organ specificity in endothelium and mesenchyme, and uncovered key ligands driving endothelial specification in the lung (e.g., WNT2B and Semaphorins) or intestine (e.g., GDF15). Upon transplantation under the kidney capsule in mice, these organoids further matured and developed perfusable human-specific sub-epithelial capillaries. Additionally, our model recapitulated the abnormal endothelial-epithelial crosstalk in patients with FOXF1 deletion or mutations. Multilineage organoids provide a unique platform to study developmental cues guiding endothelial and mesenchymal cell fate determination, and investigate intricate cell-cell communications in human organogenesis and disease. HighlightsO_LIBMP signaling fine-tunes the co-differentiation of mesoderm and endoderm. C_LIO_LIThe cellular composition in multilineage organoids resembles that of human fetal organs. C_LIO_LIMesenchyme and endothelium co-developed within the organoids adopt organ-specific characteristics. C_LIO_LIMultilineage organoids recapitulate abnormal endothelial-epithelial crosstalk in FOXF1-associated disorders. C_LI

developmental biology↗

bioRxiv2023

FOXC1 and FOXC2 maintain mitral valve endothelial cell junctions, extracellular matrix, and lymphatic vessels to prevent myxomatous degeneration

BackgroundMitral valve (MV) disease including myxomatous degeneration is the most common form of valvular heart disease with an age-dependent frequency. Genetic evidence indicates mutations of the transcription factor FOXC1 are associated with MV defects, including mitral valve regurgitation. In this study, we sought to determine whether murine Foxc1 and its closely related factor, Foxc2, are required in valvular endothelial cells (VECs) for the maintenance of MV leaflets, including VEC junctions and the stratified trilaminar extracellular matrix (ECM). MethodsAdult mice carrying tamoxifen-inducible, endothelial cell (EC)-specific, compound Foxc1;Foxc2 mutations (i.e., EC-Foxc-DKO mice) were used to study the function of Foxc1 and Foxc2 in the maintenance of mitral valves. The EC-mutations of Foxc1/c2 were induced at 7 - 8 weeks of age by tamoxifen treatment, and abnormalities in the MVs of EC-Foxc-DKO mice were assessed via whole-mount immunostaining, immunohistochemistry, and Movat pentachrome/Massons Trichrome staining. ResultsEC-deletions of Foxc1 and Foxc2 in mice resulted in abnormally extended and thicker mitral valves by causing defects in regulation of ECM organization with increased proteoglycan and decreased collagen. Notably, reticular adherens junctions were found in VECs of control MV leaflets, and these reticular structures were severely disrupted in EC-Foxc1/c2 mutant mice. PROX1, a key regulator in a subset of VECs on the fibrosa side of MVs, was downregulated in EC-Foxc1/c2 mutant VECs. Furthermore, we determined the precise location of lymphatic vessels in murine MVs, and these lymphatic vessels were aberrantly expanded in EC-Foxc1/c2 mutant mitral valves. ConclusionsOur results indicate that Foxc1 and Foxc2 are required for maintaining the integrity of the MV, including VEC junctions, ECM organization, and lymphatic vessels to prevent myxomatous mitral valve degeneration.

cell biology↗

bioRxiv2023

GENESIS CGDYN: large-scale coarse-grained MD simulation with dynamicload balancing for heterogeneous biomolecular systems

Residue-level coarse-grained (CG) molecular dynamics (MD) simulation is widely used to investigate slow biological processes that involve multiple proteins, nucleic acids, and their complexes. Biomolecules in a large simulation system are distributed non-uniformly, limiting computational efficiency with conventional methods. Here, we develop a hierarchical domain decomposition scheme with dynamic load balancing for heterogeneous biomolecular systems to keep computational efficiency even after drastic changes in particle distribution. The new schemes are applied to intrinsically disordered protein (IDP) droplet fusions. The droplet shape changes correlate with mixing IDP chains from two droplets. We also simulate formations of large IDP droplets, whose sizes are almost equivalent to those observed in microscopy. The methods have been implemented in CGDYN of the GENESIS software, which provides a new tool for investigating mesoscopic biological phenomena using the residue-level CG models.

biophysics↗

bioRxiv2023

Learning Complete Protein Representation by Deep Coupling of Sequence and Structure

Learning effective representations is crucial for understanding proteins and their biological functions. Recent advancements in language models and graph neural networks have enabled protein models to leverage primary or tertiary structure information to learn representations. However, the lack of practical methods to deeply co-model the relationships between protein sequences and structures has led to suboptimal embeddings. In this work, we propose CoupleNet, a network that couples protein sequence and structure to obtain informative protein representations. CoupleNet incorporates multiple levels of features in proteins, including the residue identities and positions for sequences, as well as geometric representations for tertiary structures. We construct two types of graphs to model the extracted sequential features and structural geometries, achieving completeness on these graphs, respectively, and perform convolution on nodes and edges simultaneously to obtain superior embeddings. Experimental results on a range of tasks, such as protein fold classification and function prediction, demonstrate that our proposed model outperforms the state-of-the-art methods by large margins.

bioinformatics↗

bioRxiv2023

Whole genome sequencing reveals autooctoploidy in the Chinese sturgeon and its evolutionary trajectories

The Order Acipenseriformes, which include sturgeons and paddlefishes, represent "living fossils" with complex genomes that are good models for understanding whole genome duplication (WGD) and ploidy evolution in fishes. Here we sequenced and assembled the first high-quality chromosome-level genome for the complex octoploid Acipenser sinensis (Chinese sturgeon), a critically endangered species that also represents a poorly understood ploidy group in Acipenseriformes. Our results show that A. sinensis is a complex autooctoploid species containing four kinds of octovalents (8 n), a hexavalent (6 n), two tetravalents (4 n), and a divalent (2 n). We propose based on an analysis taking into account delayed rediploidization that its octoploid genome composition results from two rounds of homologous whole genome duplications (WGDs), and further provide insight into the timing of its ploidy evolution. This study provides the first octoploid genome resource of Acipenseriformes for understanding ploidy composition and evolutionary trajectories of polyploidy fishes.

genomics↗

bioRxiv2023

Micelle-like clusters in phase-separated Nanog condensates: A molecular simulation study

The phase separation model for transcription suggests that transcription factors (TFs), coactivators, and RNA polymerases form biomolecular condensates around active gene loci and regulate transcription. However, the structural details of condensates remain elusive. In this study, for Nanog, a master TF in mammalian embryonic stem cells known to form protein condensates in vitro, we examined protein structures in the condensates using residue-level coarse-grained molecular simulations. Human Nanog formed micelle-like clusters in the condensate. In the micelle-like cluster, the C-terminal disordered domains, including the tryptophan repeat (WR) regions, interacted with each other near the cluster center primarily via hydrophobic interaction. In contrast, hydrophilic disordered N-terminal and DNA-binding domains were exposed on the surface of the clusters. Electrostatic attractions of these surface residues were responsible for bridging multiple micelle-like structures in the condensate. The micelle-like structure and condensate were dynamic and liquid-like. Mutation of tryptophan residues in the WR region which was implicated to be important for a Nanog function resulted in dissolution of the Nanog condensate. Finally, to examine the impact of Nanog cluster to DNA, we added DNA fragments to the Nanog condensate. Nanog DNA-binding domains exposed to the surface of the micelle-like cluster could recruit more than one DNA fragments, making DNA-DNA distance shorter. Author summaryIn eukaryotic transcription regulation, enhancer elements far from the promoter are known to modulate transcriptional activity, but the molecular mechanism remains elusive. One of these models, the phase separation model, suggests that transcription factors (TFs), coactivators, and transcription machinery form biomolecular condensates that include enhancer and promoter elements, making enhancer-promoter communication possible via the protein network in the condensate. Nanog, one of the core TFs involved in mammalian embryonic stem cells, has been reported to have the ability to form condensates. In this study, we addressed the structural details of Nanog condensates by performing residue-level coarse-grained molecular simulations. We found that Nanog formed micelle-like clusters via primarily hydrophobic interactions between tryptophan repeat regions in the C-terminal disordered domains within the condensate. On the other hand, highly charged N-terminal and DNA-binding domains were exposed to the surface of the micelle and were responsible for bridging many micelles into a condensate. The micelle-like clusters and condensate were dynamic and liquid-like. In addition, Nanog condensates could induce DNA-DNA attraction mediated by micelle-like structures.

biophysics↗

bioRxiv2023

Augmentation of a neuroprotective myeloid state by hematopoietic cell transplantation

Multiple sclerosis (MS) is an autoimmune disease associated with inflammatory demyelination in the central nervous system (CNS). Autologous hematopoietic cell transplantation (HCT) is under investigation as a promising therapy for treatment-refractory MS. Here we identify a reactive myeloid state in chronic experimental autoimmune encephalitis (EAE) mice and MS patients that is surprisingly associated with neuroprotection and immune suppression. HCT in EAE mice leads to an enhancement of this myeloid state, as well as clinical improvement, reduction of demyelinated lesions, suppression of cytotoxic T cells, and amelioration of reactive astrogliosis reflected in reduced expression of EAE- associated gene signatures in oligodendrocytes and astrocytes. Further enhancement of myeloid cell incorporation into the CNS following a modified HCT protocol results in an even more consistent therapeutic effect corroborated by additional amplification of HCT-induced transcriptional changes, underlining myeloid-derived beneficial effects in the chronic phase of EAE. Replacement or manipulation of CNS myeloid cells thus represents an intriguing therapeutic direction for inflammatory demyelinating disease.

neuroscience↗

bioRxiv2023

Surveillance of 16 UK native bat species through conservationist networks uncovers coronaviruses with zoonotic potential

There has been limited characterisation of bat-borne coronaviruses in Europe. Here, we screened for coronaviruses in 48 faecal samples from 16 of the 17 bat species breeding in the UK, collected through a bat rehabilitation and conservationist network. We recovered nine (two novel) complete genomes across six bat species: four alphacoronaviruses, a MERS-related betacoronavirus, and four closely related sarbecoviruses. We demonstrate that at least one of these sarbecoviruses can bind and use the human ACE2 receptor for infecting human cells, albeit suboptimally. Additionally, the spike proteins of these sarbecoviruses possess an R-A-K-Q motif, which lies only one nucleotide mutation away from a furin cleavage site (FCS) that enhances infectivity in other coronaviruses, including SARS-CoV-2. However, mutating this motif to an FCS does not enable spike cleavage. Overall, while UK sarbecoviruses would require further molecular adaptations to infect humans, their zoonotic risk is unknown and warrants closer surveillance.

genomics↗

bioRxiv2022

Context-dependent functions of mitochondria protein quality control in lung

Aside from its role as the universal energy source of the cell, mitochondria also control many aspects of cell behavior. In an intact tissue, whether all cells require mitochondria function to the same extent, and how mitochondria insufficiency impacts cell behavior are poorly understood. Here we show that in the mouse lung epithelium, inactivation of LONP1, an energy ATP-dependent protease that functions in the mitochondria to degrade unfolded and misfolded proteins, led to mitochondria deficiency. In the naive epithelium of the developing lung, loss of Lonp1 obliterated cell proliferation and differentiation. In the adult airway epithelium during homeostasis, loss of Lonp1 led to selective death of terminally differentiated multiciliated cells, leading to a cascade of progenitor activation to replace lost cells. In the adult airway epithelium following influenza infection, loss of Lonp1 led to failure of airway progenitor migration into the damaged alveolar region. Bulk and single cell transcriptomic analysis revealed that one branch of the ER stress pathways, namely integrated stress response (ISR), is ectopically upregulated in mutants under all three conditions. Inactivation of core ISR transcription factor ATF4 in the Lonp1 mutant airway reversed abovementioned phenotypes. Taken together, our findings demonstrate that depending on a cellular context, intact mitochondria function is required in either progenitor or progeny cells, and is essential for cell proliferation, survival or migration in the mammalian lung.

cell biology↗

bioRxiv2022

Repulsive interaction and secondary structure of highly charged proteins in regulating biomolecular condensation

Biomolecular condensation is involved in various cellular processes both functional and dysfunctional. Regulation of the condensation is thus crucial to avoid pathological protein aggregation and to maintain stable cellular environments. Recently, a class of highly charged intrinsically disordered proteins (IDPs), which are called the heat-resistant obscure (Hero) proteins, are shown to protect other client proteins from pathological aggregation. Besides the potential importance of this function, molecular mechanisms for how Hero proteins can protect other proteins from aggregation are not still known. Here we perform multiscale molecular dynamics (MD) simulations of Hero11, one of the Hero proteins, and the C-terminal region of TDP-43, as a target protein of Hero11, at various conditions to examine how they interact with each other. Based on the simulation results, three possible mechanisms have been proposed: (i) TDP-43 and Hero11 in dense phase reduces contacts with each other and shows faster diffusion due to the repulsive Hero11-Hero11 interactions, (ii) the amount of TDP-43 in dilute phase increases and their sizes become greater upon the attractive Hero11-TDP-43 interactions, and (iii) Hero-11 on the surface of small TDP-43 condensates avoids their fusions with the repulsive interactions. We also examine possible Hero-11 structures in atomistic and coarse-grained MD simulations and found disordered Hero-11 tend to assemble on the surface of the condensates, avoiding the droplet fusion effectively. The proposed mechanisms give us new insight into the regulation of biomolecular condensation in the cells and other conditions.

biophysics↗

bioRxiv2022

Inflammatory ER Stress Responses Dictate the Immunopathogenic Progression of Systemic Candidiasis

Recognition of pathogen-associated molecular patterns can trigger the IRE1 arm of the endoplasmic reticulum (ER) stress response in immune cells. IRE1 activation has been shown to maintain ER homeostasis while simultaneously coordinating diverse immunomodulatory programs in the setting of bacterial and viral infections. However, the role of IRE1 signaling in innate immune responses to fungal pathogens is unknown. Here we report that systemic infection with the fungus Candida albicans causes severe renal immunopathology by triggering inflammatory IRE1 hyperactivation in host myeloid cells. Mechanistically, sensing of fungal {beta}-glucans by the C-type lectin receptor Dectin-1 induced Src-Syk-NOX-dependent accumulation of intracellular reactive oxygen species and the ensuing generation of lipid peroxidation byproducts that sustained IRE1 activation. Selective deletion of IRE1 in leukocytes, or treatment with an IRE1 pharmacological inhibitor, reduced detrimental inflammatory responses in the kidney and extended survival in mice systemically infected with C. albicans. Hence, controlling IRE1 overactivation may be useful to impede the fatal immunopathogenic progression of disseminated candidiasis. One sentence summaryInnate IRE1 signaling in disseminated candidiasis

immunology↗