Biology subjects

Nichols, J.

Publications and source records attributed to Nichols, J..

At least 19 recordsLinked to original sources

bioRxiv2026

Pax9 governs anterior identity and deployment of sclerotome to the median fins

The caudal fin is an anomaly among vertebrate locomotory appendages: its internal skeleton is as asymmetric as the human hand, but it lacks the Shh-secreting Zone of Polarizing Activity (ZPA) and the Gli3/HoxD/Hand2 programs that govern anterior-posterior patterning in other appendages. As the caudal fin was also the first appendage to evolve, deciphering its alternative patterning program may provide clues to the ancestral state. Pax9 is one of few conserved appendage patterning factors reported to also be active in the caudal fin, specifically in the anterior domain. We report that loss of pax9 function in zebrafish not only disrupts anterior-specific caudal fin anatomy, but also results in a spectacular fusion of the caudal and anal fins along the ventral midline. The dorsal fin is also expanded to a lesser degree, the paired fins not at all. The mutant caudal fin initially forms as an irregularly patterned structure lacking anterior molecular identity, with supernumerary elements spilling out beyond the normal anterior boundary. Unexpectedly, this phenotype is subsequently compounded by neighboring trunk somites erroneously deploying skeletal mesenchyme in the normally finless caudal peduncle region, completing the ectopic skeleton. scRNAseq analysis at caudal fin bud stage indicates that pax9 mutants gain skeletal mesenchyme at the expense of a specialized type of fibroblast involved in the formation of fin fold actinotrichia. Median fin skeletal mesenchyme and fin fold fibroblasts both arise from the sclerotome, a somite compartment that also robustly expresses pax9. We propose that, within the sclerotome, Pax9 pushes progenitors towards fin fold fibroblast fate, limiting how many cells will later be available to make median fin skeleton. Within the fin bud, it drives anterior identity, with the strongest impact on the ZPA-free caudal fin bud. These dual sites of action make Pax9 a uniquely powerful governor of median fin development.

Developmental Biology↗

bioRxiv2026

Array-CNCC: precise aggregation and arrayed plating facilitate quantitative phenotyping of human cranial neural crest cells and craniofacial disease modelling

Facial development is highly sensitive to genetic and environmental perturbation, with craniofacial malformation associated with over one-third of congenital birth defects. The face arises during an early and largely inaccessible window of embryonic development, with a large contribution from transient and multipotent cranial neural crest cells (CNCCs). Assessment of the molecular and cellular mechanisms driving normal and disordered human facial development therefore relies greatly on the use of in vitro cellular models. Here, we adapted a neurosphere-based CNCC differentiation protocol to facilitate robust quantification of early specification and migration events. Introduction of single-cell aggregation with arrayed plating enabled standardisation of neurosphere size, growth and patterning. Inclusion of fibronectin coating enhanced the efficiency of neurosphere attachment and synchronicity of CNCC migration timing. To demonstrate application of the Array-CNCC method, we developed a strategy for mosaic co-culture, which can facilitate differentiation of wildtype untreated cells directly alongside cells exposed to distinct drug treatments or genetic alterations. Finally, we present a screening approach which we use to test the impact of distinct extracellular matrix components on neurosphere morphology, CNCC migration and gene expression. Together, the Array-CNCC method is highly amenable to quantitative phenotyping and screening approaches, enabling enhanced craniofacial disease modelling with both cellular and molecular readouts.

developmental biology↗

bioRxiv2025

Simplified In Vitro Generation of Human Gastruloids for Modelling Early Development

The purpose of this study was to optimize the efficiency and cost of human gastruloid formation by testing and adjusting individual parameters using as examples two distinct human pluripotent stem cell lines, both available from the UK Stem Cell Bank. For the first step, commercially sourced culture medium was replaced with a home-made defined recipe, known as N2B27, into which specific reagents can be titrated. By reducing the concentration of Activin A to 15% of the original protocol, efficient elongation of aggregated embryoid bodies was achieved. Also, titrating initial cell density and delaying the brief culture in GSK3 inhibitor until the onset of cell aggregation in individual wells was advantageous. Efficiency of formation of early gastruloids exhibiting the expected regionalization of the three embryonic germ layers was further enhanced by addition of TGF{beta}-inhibitor. The optimization steps presented here thus provide a simplified, robust and relatively economical protocol for consistent generation of elongated gastruloids from human pluripotent stem cells. This streamlined method improves accessibility and reproducibility, also providing a standardized platform to investigate fundamental principles of early human development. Summary StatementWe present an optimized protocol for human gastruloid production that should enhance efficiency, reproducibility and affordability for future in vitro studies into early human post-implantation development.

developmental biology↗

bioRxiv2025

QSProteome: A Community-Driven Interactive Platform for Large-Scale Exploration and Evaluation of Predicted Protein Complex Structures.

QSProteome (https://QSProteome.org) is a community-scale platform for modeling, evaluating, and refining quaternary protein structure. The resource hosts 35,528 unique modeled assemblies spanning over 42,000 genes, covering nearly all curated complexes in BioCyc and ComplexPortal databases. Each model is displayed on an interactive page with 3D visualization, chain-level confidence metrics, structural alignments, functional annotations, and automated stoichiometry checks against database expectations. A cloud-based server supports continuous user uploads and automated processing pipelines--enabling submission and validation of >54,000 models within 14 weeks. To promote iterative refinement, QSProteome includes a gamified re-curation workflow that transitions users from training modules into live curation, enabling the community-led assessment of 1,547 ABC transporter complexes. Together, these components form a dynamic, scalable infrastructure for proteome-scale structural biology. By unifying modeling, validation, and annotation in a reusable, searchable, and community-extensible framework, QSProteome enables proteome-scale structure accessibility and reuse--powering discovery, annotation, and collaborative refinement across the structural biology community.

bioinformatics↗

bioRxiv2025

A new resource of clonal pluripotent human stem cell lines exhibiting inter- and intra-embryo consistency and variability

Human naive pluripotent stem cells can generate all somatic tissues and extra-embryonic components of the blastocyst. We derived multiple clonal naive pluripotent stem cell lines from individual embryos by physical separation of inner cell mass cells and subsequent individual expansion of each resulting dome-shaped colony, providing the foundation for a resource to investigate intra- and inter-embryo variation. Twenty lines were derived from ten embryos donated from nine couples. While differences between lines are observed, the overarching pluripotency circuitry is preserved in each. They can differentiate into extra-embryonic lineages and readily acquire post-implantation pluripotent identity when exposed to culture conditions driving in vitro capacitation, subsequently to generate derivatives of the three germ layers: ectoderm, mesoderm and endoderm. Some lines exhibit intra-chromosomal amplification and deletions and are therefore anticipated to provide a valuable, accessible system for modelling chromosomal mosaicism and its potential consequences using chimeric organoids, such as blastoids and gastruloids.

developmental biology↗

bioRxiv2025

Quantifying virus load and characterising virus diversity in wildlife samples with target enrichment sequencing

Metagenomics is a powerful tool for characterising viruses, with broad applications across diverse disciplines, from understanding the ecology and evolutionary history of viruses to identifying causative agents of emerging outbreaks with unknown aetiology. Additionally, metagenomic data contains valuable information about the amount of virus present within samples. However, we have yet to leverage metagenomics to assess viral load, which is a key epidemiological parameter. To effectively use sequencing outputs to inform transmission, we need to understand the relationship between read depth and viral load across a diverse set of viruses. Here, using target enrichment sequencing, we investigated the detection and recovery of virus genomes by spiking known concentrations of DNA and RNA viruses into wild rodent faecal samples. In total, 15 experimental replicates were sequenced with target enrichment sequencing and compared to shotgun sequencing of the same background samples. Target enriched sequencing recovered all spike-in viruses at every concentration (102, 103, and 105 {+/-} 1 log genome copies) and showed a log-linear relationship between spike-in concentration and mean read depth. Background viruses (including Kobuvirus and Cardiovirus) were recovered consistently across all biological and technical replicates, but genome coverage was variable between virus genera and likely reflected the composition of target enrichment probe panel. Overall, our study highlights the strengths and weaknesses of using commercially available panels to quantify and characterise wildlife viromes, and underscores the importance of probe panel design for accurately interpreting coverage and read depth. To advance the use of metagenomics for understanding virus transmission, further research will be needed to elucidate how sequencing strategy (e.g. library depth, pooling), virome composition, and probe design influence viral read counts and genome coverage.

genomics↗

bioRxiv2025

Fitness Landscape Ruggedness Arises from Biophysical Complexity

The ruggedness of a proteins fitness landscape constrains how it can evolve and how effectively it can be engineered. What determines whether a fitness landscape is rugged, and where in sequence space that ruggedness concentrates, remains poorly understood. Existing measures of ruggedness rely on complete combinatorial datasets and cannot be applied to landscapes in which only a fraction of possible sequences have been characterized. Here, we introduce a graph-based measure of ruggedness based on how quickly fitness differences diffuse across a landscape, which can be applied to sparse datasets. Applying this measure to 64 protein domains and several experimental systems with characterized folding and binding constraints, we show that ruggedness is not uniformly distributed but concentrates at regions where the dominant constraint on fitness changes. These results establish that the composition of individually simple constraints can generate localized ruggedness, providing a mechanistic basis for the complexity of protein fitness landscapes.

evolutionary biology↗

bioRxiv2024

A Spatiotemporal Atlas of Mouse Gastrulation and Early Organogenesis to Explore Axial Patterning and Project In Vitro Models onto In Vivo Space

At the onset of murine gastrulation, pluripotent epiblast cells migrate through the primitive streak, generating mesodermal and endodermal precursors, while the ectoderm arises from the remaining epiblast. Together, these germ layers establish the body plan, defining major body axes and initiating organogenesis. Although comprehensive single cell transcriptional atlases of dissociated mouse embryos across embryonic stages have provided valuable insights during gastrulation, the spatial context for cell differentiation and tissue patterning remain underexplored. In this study, we employed spatial transcriptomics to measure gene expression in mouse embryos at E6.5 and E7.5 and integrated these datasets with previously published E8.5 spatial transcriptomics1 and a scRNA-seq2 atlas spanning E6.5 to E9.5. This approach resulted in a comprehensive spatiotemporal atlas, comprising over 150,000 cells with 88 refined cell type annotations as well as genome-wide transcriptional imputation during mouse gastrulation and early organogenesis. The atlas facilitates exploration of gene expression dynamics along anterior-posterior and dorsal-ventral axes at cell type, tissue, and organismal scales, revealing insights into mesodermal fate decisions within the primitive streak. Moreover, we developed a bioinformatics pipeline to project additional scRNA-seq datasets into a spatiotemporal framework and demonstrate its utility by analysing cardiovascular models of gastrulation3. To maximise impact, the atlas is publicly accessible via a user-friendly web portal empowering the wider developmental and stem cell biology communities to explore mechanisms of early mouse development in a spatiotemporal context.

developmental biology↗

bioRxiv2024

Molecular cloning and host range analysis of three cytomegaloviruses from Mastomys natalensis

Herpesvirus-based vectors are attractive for use as conventional or transmissible vaccines against emerging zoonoses in inaccessible animal populations. In both cases, cytomegaloviruses as members of the subfamily Betaherpesvirinae are particularly suitable for vaccine development as they are highly specific for their natural host species, infect a large proportion of their host population, and cause mild infections in healthy individuals. The Natal multimammate mouse (Mastomys natalensis) is the natural reservoir of Lassa virus, which causes deadly hemorrhagic fever in humans. M. natalensis was recently reported to harbor at least three different cytomegaloviruses (MnatCMV1, MnatCMV2 and MnatCMV3). Herein, we report the molecular cloning of three complete MnatCMV genomes in a yeast and bacterial artificial chromosome (YAC-BAC) hybrid vector. Purified viral genomes were cloned in yeast by single-step transformation-associated recombination (STAR cloning) and subsequently transferred to Escherichia coli for further genetic manipulation. Integrity of the complete cloned viral genomes was verified by sequencing, and replication fitness of viruses reconstituted from these clones was analyzed by replication kinetics in M. natalensis fibroblasts and kidney epithelial cells. We also found that neither parental nor cloned MnatCMVs replicated in mouse and rat fibroblasts, nor did they show sustained replication in baby hamster kidney cells, consistent with the expected narrow host range for these viruses. We further demonstrated that an exogenous sequence can be inserted by BAC-based mutagenesis between open reading frames M25 and m25.1 of MnatCMV2 without affecting replication fitness in vitro, identifying this site as potentially suitable for the insertion of vaccine target antigen genes. ImportanceCytomegaloviruses recently discovered in the Natal multimammate mouse (Mastomys natalensis) are widespread within the M. natalensis population. Since these rodents also serve as natural hosts of the human pathogen Lassa virus (LASV), we investigated the potential suitability of M. natalensis CMVs (MnatCMVs) as vaccine vectors. We describe the cloning of three different MnatCMV genomes as bacterial artificial chromosomes (BACs). Replicative capacity and species specificity of these BAC-derived MnatCMVs were analyzed in multiple cell types. We also identified a transgene insertion site within one of the MnatCMV genomes suitable for the incorporation of vaccine target antigens. Together, this study provides a foundation for the development of MnatCMVs as transmissible MnatCMV-based LASV vaccines to reduce LASV prevalence in hard-to-reach M. natalensis populations and thereby zoonotic transmission to humans.

microbiology↗

bioRxiv2024

Donor embryonic stem cells impede host epiblast specification in 8-cell stage chimeras by crowding and FGF4 signalling

During mouse embryo compaction, outer cells become trophectoderm, while inner cells form the inner cell mass (ICM), later differentiating into primitive endoderm and epiblast - source of embryonic stem cells (ESCs) - during blastocyst formation. Trophectoderm specification is driven by position-governed polarisation, while primitive endoderm specification is positively regulated by FGF4 signalling from the ICM and epiblast. When injected into an 8-cell stage morula, ESCs can exclude host cells from the epiblast, leading to mice derived entirely from these cells. While evidence suggests roles for ESC-produced FGF4 and physical crowding in host cell displacement from the ICM, the interplay between these possible mechanisms has yet to be dissected, in part due to the lack of studies using Fgf4-/- ESCs. Here, we combine chimaera titration assays with mathematical modelling to study these mechanisms of host cell displacement. Both Fgf4+/+ and Fgf4-/- ESCs displaced host cells from the epiblast, while only Fgf4-/- injected embryos reduced primitive endoderm and increased trophectoderm, indicating sequential exclusion by displacement crowding followed by FGF4 signalling. SUMMARY STATEMENTIn mouse blastocyst chimeras, donor cells displace host cells from the epiblast by a combination of crowding and FGF4 signalling.

developmental biology↗

bioRxiv2024

Genetic diversity, population structure and differentiation of farmed and wild African catfish (Clarias gariepinus) in Nigeria

The African catfish (Clarias gariepinus) is a commercially important species, for both fisheries and aquaculture, and is now the most commonly farmed fish in sub-Saharan Africa. However, knowledge about the genetic diversity and population structure of wild and farmed populations, which is crucial for effective conservation and sustainable aquaculture management, is scarce. Using mitochondrial DNA (mtDNA) cytochrome c oxidase 1 gene (COI) sequencing and genomic analysis using triple restriction site-associated DNA sequencing (3RAD), we investigated the genetic diversity and population structure of farmed and wild C. gariepinus populations from Nigeria, including an albino form found in the wild. Eleven COI haplotypes were identified, of which seven were unique to wild samples. Wild sampling sites had a slightly broader range and higher maximum values for observed heterozygosity (Ho = 0.109 - 0.165), expected heterozygosity (He = 0.111 - 0.216), and nucleotide diversity (pi = 0.125 - 0.225) compared to the farmed populations (Ho = 0.118 - 0.147, He = 0.112-0.144, pi = 0.117 - 0.151). Conversely, genetic differentiation (Fst) was higher among farmed sampling sites compared to the wild ones and there was high genetic differentiation between the farmed and wild C. gariepinus sampling sites (Fst = 0.31 - 0.47). Despite evidence for admixture for both farmed and wild fish, there was little evidence of admixture between the two groups. Nevertheless, both mtDNA and 3RAD data strongly suggested that the albino fish, collected from the wild, were in fact farm escapees. Despite overall differentiation farmed genotypes suggesting that overall genetic integrity of wild fish has been maintained, this evidence of escape provides a warning about potential risks of increasing aquaculture activities. Specifically, this indicates the need for greater regulation of fish farms to monitor and reduce the risk of escapes.

genomics↗

bioRxiv2024

WNT-mediating TCF/LEF transcription factor gene expression in early human pluripotency and cell lineages differs from the rodent paradigm

Embryonic stem cell research has uncovered different requirements for WNT/{beta}-catenin signalling in human naive pluripotent cells compared to the mouse paradigm. It is therefore important to study WNT/{beta}-catenin signalling directly in models of early human development. Since TCF/LEF factors mediate the regulation of target genes downstream of WNT/{beta}-catenin signalling, we studied the expression and protein localisation of the four TCF/LEF genes by analysing in vitro "snapshots" of human development, leveraging naive and primed pluripotent cells as well as extraembryonic and early embryonic cell lineages. Strikingly, we comprehensively confirm clear differences between mouse and human pluripotent stem cells, suggesting species-specific requirements for WNT signalling that may reflect differences in states of pluripotency. Human naive ES cells express very low TCF7L1, unlike their mouse counterparts. TCF7L2 is robustly expressed in human naive ES-derived trophectoderm cells. In human primed pluripotent stem cells, activation of WNT/{beta}-Catenin signalling is required to induce expression of both TCF7 and LEF1, concomitant with hallmark gastrulation markers. This expression of human TCF/LEF genes benchmarks differential requirements for WNT/{beta}-catenin signalling throughout early human embryo development that requires further investigation.

developmental biology↗

bioRxiv2024

STAT3 signalling enhances tissue expansion during postimplantation mouse development

STAT3 signalling has been studied extensively in the context of self-renewal and differentiation of mouse embryonic stem cells. Zygotic STAT3 is required for normal postimplantation development. On an outbred genetic background, Stat3 null embryos consistently lagged behind their littermates, beginning with significant reduction of epiblast cells at implantation. Remarkably, mutants closely resemble non-affected embryos from the previous day at all postimplantation stages examined. We pinpoint this phenotype to loss of the serine-phosphorylated form of STAT3 which predominates in postimplantation embryonic tissues. Bulk RNA-sequencing analysis of isolated mouse epiblasts confirmed Stat3 null embryos exhibited developmental delay transcriptionally. Single cell RNA sequencing of mid gestation chimaeras containing STAT3 null embryonic stem cells revealed exclusion of mutant cells exclusively from the erythroid lineage. Although Stat3 null embryonic stem cells can differentiate into erythroid and hematopoietic lineages in vitro, they are out-competed when mixed with wild type cells. Combined with the reduced size of STAT3 null epiblasts after implantation, our results implicate a role for STAT3 in cell proliferation affecting temporal control of embryonic progression and rapid differentiation. For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission.

developmental biology↗

bioRxiv2024

Epigenetic priming of embryonic enhancer elements coordinates developmental gene networks

Embryonic development requires the accurate spatiotemporal execution of cell lineage-specific gene expression programs, which are controlled by transcriptional enhancers. Developmental enhancers adopt a primed chromatin state prior to their activation; however how this primed enhancer state is established, maintained, and how it affects the regulation of developmental gene networks remains poorly understood. Here, we use comparative multi-omic analyses of human and mouse early embryonic development to identify subsets of post-gastrulation lineage-specific enhancers which are epigenetically primed ahead of their activation, marked by the histone modification H3K4me1 within the epiblast. We show that epigenetic priming occurs at lineage-specific enhancers for all three germ layers, and that epigenetic priming of enhancers confers lineage-specific regulation of key developmental gene networks. Surprisingly in some cases, lineage-specific enhancers are epigenetically marked already in the zygote, weeks before their activation during lineage specification. Moreover, we outline a generalisable strategy to use naturally occurring human genetic variation to delineate important sequence determinants of primed enhancer function. Our findings identify an evolutionarily conserved program of enhancer priming and begin to dissect the temporal dynamics and mechanisms of its establishment and maintenance during early mammalian development.

developmental biology↗

bioRxiv2024

Comparative study of two Rift Valley fever virus field strains circulating in Mauritania in 2010 and 2013 reveals the high virulence of the MRU25010-30 strain isolated from camel

Rift Valley fever (RVF) is one of the major viral arthropod-borne diseases in Africa. In recent decades, RVF virus (RVFV), the causative agent of RVF, has been responsible for multiple outbreaks in West Africa with important consequences on human and animal health. In particular, an outbreak occurred in 2010 after heavy rainfalls in the desertic region of Adrar, Mauritania. It was characterized by the appearance of severe clinical signs among dromedary camels. Another one occurred in 2013-2014 across Senegal and the southern part of Mauritania. In this study, we characterized two RVFV field strains isolated during these two outbreaks. The first strain, MRU25010-30, has been isolated in camel (2010) while the second, MRU2687-3, was isolated in goat (2013). By deep-sequencing and rapid amplification of cDNA-ends by polymerase chain reaction (RACE-PCR), we successfully sequenced the complete genome of these two RVFV strains as well as the reference laboratory strain ZH548. Phylogenetic analysis shows that the two field viruses belong to two different RVFV genetic lineages. Moreover, we show that MRU25010-30 replicates more efficiently in various in vitro cell culture models than MRU2687-3 and ZH548. In vivo, MRU25010-30 caused rapid death of BALB/c mice and proved to be more virulent than MRU2687-3, regardless of the route of inoculation (subcutaneous or intranasal). The virulence of MRU25010-30 is associated with a high viral load in the liver and serum of infected mice, while the death of mice infected with MRU2687-3 and ZH548 correlates with a high viral load in the brain. Altogether, the data presented in this study provide new avenues to unveil the molecular viral determinants that modulate RVFV virulence and replication capacity Author SummaryRift Valley fever is an arboviral zoonosis caused by Rift Valley fever virus (RVFV) belonging to the Phlebovirus genus. It poses a major risk for causing a public and animal health emergency and is a significant economic burden in many African countries. To date, our knowledge of the impact of RVFV genetic diversity on its virulence, replicative capacities and transmission by mosquitoes is limited. In this study, we fully sequenced two RVFV strains isolated in Mauritania during two distinct outbreaks (2010 and 2013) and show that they were genetically distant. Interestingly, we show that one of the strains (MRU25010-30) is able to replicate in vitro more efficiently than the other (MRU2687-3). Additionally, we show that high levels of viremia and viral load in the liver are associated with rapid death in BALB/c mice infected with MRU25010-30, whereas mice infected by MRU2687-3 tend to die later with high viral load in the brain. In conclusion, our study confirms that RVFV strains from distinct genetic lineages have different phenotypic characteristics such as virulence and replication capacity. These data provide a strong basis for further studies aimed at identifying the viral genetic determinants responsible for the observed phenotypes.

microbiology↗

bioRxiv2024

A Machine Learning Framework to Identify the Correlates of Disease Severity in Acute Arbovirus Infection

Most viral diseases display a variable clinical outcome due to differences in virus strain virulence and/or individual host susceptibility to infection. Understanding the biological mechanisms differentiating a viral infection displaying severe clinical manifestations from its milder forms can provide the intellectual framework toward therapies and early prognostic markers. This is especially true in arbovirus infections, where most clinical cases are present as mild febrile illness. Here, we used a naturally occurring vector-borne viral disease of ruminants, bluetongue, as an experimental system to uncover the fundamental mechanisms of virus-host interactions resulting in distinct clinical outcomes. As with most viral diseases, clinical symptoms in bluetongue can vary dramatically. We reproduced experimentally distinct clinical forms of bluetongue infection in sheep using three bluetongue virus (BTV) strains (BTV-1IT2006, BTV-1IT2013 and BTV-8FRA2017). Infected animals displayed clinical signs varying from clinically unapparent, to mild and severe disease. We collected and integrated clinical, haematological, virological, and histopathological data resulting in the analyses of 332 individual parameters from each infected and uninfected control animal. We subsequently used machine learning to identify the key viral and host processes associated with disease pathogenesis. We identified five different fundamental processes affecting the severity of bluetongue: (i) virus load and replication in target organs, (ii) modulation of the host type-I IFN response, (iii) pro-inflammatory responses, (iv) vascular damage, and (v) immunosuppression. Overall, our study using an agnostic machine learning approach, can be used to prioritise the different pathogenetic mechanisms affecting the disease outcome of an arbovirus infection.

microbiology↗

bioRxiv2023

Leveraging ancestral sequence reconstruction for protein representation learning

Protein language models (PLMs) convert amino acid sequences into the numerical representations required to train machine learning (ML) models. Many PLMs are large (>600 M parameters) and trained on a broad span of protein sequence space. However, these models have limitations in terms of predictive accuracy and computational cost. Here, we use multiplexed Ancestral Sequence Reconstruction (mASR) to generate small but focused functional protein sequence datasets for PLM training. Compared to large PLMs, this local ancestral sequence embedding (LASE) produces representations 10-fold faster and with higher predictive accuracy. We show that due to the evolutionary nature of the ASR data, LASE produces smoother fitness landscapes in which protein variants that are closer in fitness value become numerically closer in representation space. This work contributes to the implementation of ML-based protein design in real-world settings, where data is sparse and computational resources are limited.

bioinformatics↗

bioRxiv2023

Human naive stem cell models reveal the role of FGF signalling in hypoblast specification in the human embryo

The hypoblast is an essential extra-embryonic tissue set aside within the inner cell mass early in mammalian embryo development, in the blastocyst. Research with human embryos is challenging. Thus, stem cell models that reproduce hypoblast differentiation provide valuable alternatives. We show here that human naive PSC to hypoblast differentiation proceeds via reversion to a transitional ICM-like state, from which hypoblast emerges in concordance with the trajectory in human blastocysts. We identified a window when fibroblast growth factor (FGF) signalling is critical for hypoblast specification. Revisiting FGF signalling in human embryos revealed that inhibition in the early blastocyst suppresses hypoblast formation. In vitro, the induction of hypoblast is synergistically enhanced by limiting trophectoderm and epiblast fates. This finding revises previous reports and establishes a conservation in lineage specification between mouse and human. Overall, this study demonstrates the utility of human naive PSC-based models in elucidating mechanistic features of early human embryogenesis.

developmental biology↗