Biology subjects

Qiang, B.

Publications and source records attributed to Qiang, B..

6 recordsLinked to original sources

bioRxiv2025

De novo Design of All-atom Biomolecular Interactions with RFdiffusion3

Deep learning has accelerated protein design, but most existing methods are restricted to generating protein backbone coordinates and often neglect interactions with other biomolecules. We present RFdiffusion3 (RFD3), a diffusion model that generates protein structures in the context of ligands, nucleic acids and other non-protein constellations of atoms. Because all polymer atoms are modeled explicitly, conditioning the model on complex sets of atom-level constraints for enzyme design and other challenges is both simpler and more effective than previous approaches. RFD3 achieves improved performance compared to prior approaches on a range of in silico benchmarks with one tenth the computational cost. Finally, we demonstrate the broad applicability of RFD3 by designing and experimentally characterizing DNA binding proteins and cysteine hydrolases. The ability to rapidly generate protein structures guided by complex sets of atom-level constraints in the context of arbitrary non-protein atoms should further expand the range of functions attainable through protein design.

biochemistry↗

bioRxiv2025

Growth Factor-Based Manufacturing of Human Pluripotent Stem Cell-Derived Cardiomyocytes Using the Vertical Wheel Bioreactor System

BackgroundMultiple protocols have been reported for the large-scale generation of human pluripotent stem cell-derived cardiomyocytes (hPSC-CMs) in bioreactors using small molecules; however, no comparable bioreactor-based methods have been established using growth factors. This is despite evidence that differentiation with optimized concentrations of BMP4, Activin A, and bFGF offers finer control of cardiomyocyte phenotype. Here, we develop scalable hPSC expansion and growth factor-based cardiac differentiation protocols using the vertical wheel bioreactor (VWBR) system. Methods and ResultsThe expansion of undifferentiated hPSCs was first optimized in 100 mL VWBRs by varying parameters, including starting cell seeding density, agitation rate, and media exchange schedule. Proliferation, viability, aggregate diameter, media metabolites, and pluripotency were assessed during hPSC expansion. Next, we evaluated the effects of undifferentiated hPSC culture conditions on subsequent cardiomyocyte differentiation potential. We found that hPSCs expanded in static culture or in VWBRs at different densities and agitation rates all differentiated into hPSC-CM populations of similar cardiac purity; however, cardiomyocyte yields were initially lower when VWBR-expanded hPSCs were used. We compared the differentiation kinetics of hPSCs expanded in VWBRs to conventional 2D culture and found that the former had accelerated mesodermal commitment and significantly greater cKit+/CXCR4+/PDGRF- cell formation during differentiation. Modifying our aggregation and mesoderm induction steps improved cell yields and enabled reliable production of >1x106 cells/mL cardiac troponin T+ (cTnT) hPSC-CMs. Highlighting the versatility of our growth factor-based system, variation in the BMP4:Activin A ratio enabled a second heart field-like differentiation and generation of atrial-like cardiomyocytes in VWBRs. We further show that our expansion and differentiation protocols are reproducible and economical in 500 mL VWBRs, yielding on average 1.11x106 hPSC-CMs/mL at a mean purity of 93% cTnT+. Characterization of VWBR produced hPSC-CM force generation, action potentials, and intercellular calcium transients confirmed the expected phenotype of ventricular-like cells. Lastly, VWBR produced hPSC-CMs robustly engrafted in the infarcted guinea pig myocardium, supporting use as a cell therapy product. ConclusionsThis novel bioreactor-based protocol will enable cardiac cell therapy and tissue engineering applications by providing scalable and consistent production of hPSC-derived cardiac cell products.

cell biology↗

bioRxiv2025

Human pluripotent stem cell-derived atrioventricular node-like pacemaker cells exhibit biological conduction bridge properties in vitro and in vivo

The atrioventricular node (AVN) ensures synchronized heart contractions by establishing the electrical connection between the atria and ventricles. Dysfunction of the pacemaker cells of the AVN leads to atrioventricular block (AV block), a life-threatening condition managed with electronic pacemakers (EPMs). EPMs have drawbacks that could be overcome by a human pluripotent stem cell (hPSC)-derived biological conduction bridge (BioCB). Recent studies demonstrated the differentiation of AVN-like cells from hPSCs, but their conduction properties upon engraftment in vivo remain unexplored. Here we report the generation of AVN-like pacemaker cells (AVNLPCs) from hPSCs using WNT and BMP signaling modulation. These AVNLPCs transcriptionally resemble fetal AVN pacemaker cells, exhibit pacemaker action potentials, and display unique AVN-like conduction properties. Notably, when transplanted into the guinea pig heart, AVNLPCs replicate the functional properties of the AVN. Our study highlights the potential of an AVNLPC-based BioCB as novel cell therapy to improve treatment for AV block patients.

developmental biology↗

bioRxiv2025

Accelerating Biomolecular Modeling with AtomWorks andRF3

Deep learning methods trained on protein structure databases have revolutionized biomolecular structure prediction, but developing and training new models remains a considerable challenge. To facilitate the development of new models, we present AtomWorks: a broadly applicable data framework for developing state-of-the-art biomolecular foundation models spanning diverse tasks, including structure prediction, generative protein design, and fixed backbone sequence design. We use AtomWorks to train RosettaFold-3 (RF3), a structure prediction network capable of predicting arbitrary biomolecular complexes with an improved treatment of chirality that narrows the performance gap between closed-source AlphaFold3 (AF3) and existing open-source implementations. We expect that AtomWorks will accelerate the next generation of open-source biomolecular machine learning models and that RF3 will be broadly useful as a structure prediction tool. To this end, we release the AtomWorks framework (https://github.com/RosettaCommons/atomworks), together with curated training data, code and model weights for RF3 (https://github.com/RosettaCommons/modelforge) under a permissive BSD license.

biochemistry↗

bioRxiv2023

Temporal transcriptomic dynamics in developing macaque neocortex

Despite intense research on mice, the transcriptional regulation of neocortical neurogenesis remains limited in humans and non-human primates. Cortical development in rhesus macaque is known to recapitulate multiple facets of cortical development in humans, including the complex composition of neural stem cells and the thicker supragranular layer. To characterize temporal shifts in transcriptomic programming responsible for differentiation from stem cells to neurons, we sampled parietal lobes of rhesus macaque at E40, E50, E70, E80, and E90, spanning the full period of prenatal neurogenesis. Single-cell RNA sequencing produced a transcriptomic atlas of the developing rhesus macaque neocortex parietal lobe. Identification of distinct cell types and neural stem cells emerging in different developmental stages revealed a terminally bifurcating trajectory from stem cells to neurons. Notably, deep-layer neurons appear in the early stages of neurogenesis, while upper-layer neurons appear later. While these different lineages show overlap in their differentiation program, cell fates are determined post-mitotically. Pseudotime trajectories from vRGs to oRGs revealed differences in dynamic gene expression profiles and identified divergence in their activation of BMP, FGF, and WNT signaling pathways. These results provide a comprehensive picture of the temporal patterns of gene expression leading to different fates of radial glial progenitors during neocortex layer formation.

developmental biology↗

bioRxiv2018

COMPASS Family Histone Methyltransferase ASH2L Mediates Corticogenesis via Transcriptional Regulation of Wnt Signalling

Cell fate specification in neural progenitor cells (NPCs) is orchestrated via extrinsic and intrinsic molecular programs, and histone methylation in these decisions has been ascribed to a crucial function regulating gene expression. Here, we show that the COMPASS family histone methyltransferase co-factor ASH2L is required in NPCs proliferation and upper layer cortical projection neurons production and position. Deletion of Ash2l impairs trimethylation of H3K4 and transcriptional machinery specifically for subsets of Wnt-{beta}-catenin signalling, disrupting their transcription and consequently inhibiting the proliferation ability of NPCs in late stages of neurogenesis. Consistently, Ash2l conditional mutants exhibit thinning neocortex with reduced upper layer neurons and altered neuronal position. Moreover, overexpressing {beta}-catenin after Ash2l elimination or knockdown can rescue the proliferation deficiency of NPCs both in vivo and in vitro. These results demonstrate an essential and highly specific role for Ash2l in controlling NPCs proliferation and late-born neurons lamination in corticogenesis via transcriptionally regulating Wnt-{beta}-catenin signalling, and provide clues to how the COMPASS family epigenetic factors coordinate cell fate determination during cortex development.

neuroscience↗