UHRF1P contributes to IL-17A-mediated systemic lupus erythematosus via UHRF1-MAP4K3 axis.

Inflammatory T cells contribute to the pathogenesis of autoimmune diseases such as systemic lupus erythematosus (SLE). Analysis of the T-cell transcriptomics data of two independent SLE patient cohorts by three machine learning models revealed the pseudogene UHRF1P as a novel SLE biomarker. The pseudogene-encoded UHRF1P protein was overexpressed in peripheral blood T cells of SLE patients. The UHRF1P protein lacks the amino-terminus of its parental UHRF1 protein, resulting in missing the proteasome-binding ubiquitin-like (Ubl) domain of UHRF1. T-cell-specific UHRF1P transgenic mice manifested the induction of IL-17A and autoimmune inflammation. Mechanistically, UHFR1P prevented UHRF1-induced Lys48-linked ubiquitination and degradation of MAP4K3 (GLK), which is a kinase known to induce IL-17A. Consistently, IL-17A induction and autoimmune phenotypes of UHRF1P transgenic mice were obliterated by MAP4K3 knockout. Collectively, UHRF1P overexpression in T cells inhibits the E3 ligase function of its parental UHRF1 and induces autoimmune diseases.

Convergent evolution of kidney sizes and supraorbital salt glands for birds living in saline habitats.

Only a small number of avian species inhabit salty environments. To understand how they adapted, we examined the evolution of kidney sizes, supraorbital salt glands (SSGs), and the utilization of salty habitats across 230 species spanning 25 avian orders. Phylogenetic analysis indicates that SSGs, large kidneys, and thriving in salty habitats emerged convergently in birds. Transition rate analysis reveals that species possessing SSGs and large kidneys tended to move from low-to high-salinity environments, while others moved in the opposite direction. However, habitat salinity also influenced kidney evolution; lineages residing in high-salinity environments tended to develop larger kidneys than those in low-salinity environments. Our findings suggest that SSGs and large kidneys may have evolved through adaptation to high salinity. Overall, habitat conditions and physiological traits influenced avian adaptation to salty environments in a reciprocal manner. These results shed the new light on the evolutionary mechanisms underlying functional diversity in birds.

Degeneration of the Olfactory System in a Murid Rodent that Evolved Diurnalism.

In mammalian research, it has been debated what can initiate an evolutionary tradeoff between different senses, and the phenomenon of sensory tradeoff in rodents, the most abundant mammalian clade, is not evident. The Nile rat (Arvicanthis niloticus), a murid rodent, recently adapted to a diurnal niche through an evolutionary acquisition of daylight vision with enhanced visual acuity. As such, this model provides an opportunity for a cross-species investigation where comparative morphological and multi-omic analyses of the Nile rat are made with its closely related nocturnal species, e.g. the mouse (Mus musculus) and the rat (Rattus norvegicus). Thus, morphological examinations were performed, and evolutionary reductions in relative sizes of turbinal bone surfaces, the cribriform plate, and the olfactory bulb were discovered in Nile rats. Subsequently, we compared multiple murid genomes, and profiled olfactory epithelium transcriptomes of mice and Nile rats at various ages with RNA sequencing. The results further demonstrate that, in comparison with mouse olfactory receptor (OR) genes, Nile rat OR genes have experienced less frequent gain, more frequent loss, and more frequent expression reduction during their evolution. Furthermore, functional degeneration of coding sequences in the Nile rat lineage was found in OR genes, yet not in other genes. Taken together, these results suggest that acquisition of improved vision in the Nile rat has been accompanied by degeneration of both olfaction-related anatomical structures and OR gene repertoires, consistent with the hypothesis of an olfaction-vision tradeoff initiated by the switch from a nocturnal to a diurnal lifestyle in mammals.

Dysfunction of duplicated pair rice histone acetyltransferases causes segregation distortion and an interspecific reproductive barrier.

Postzygotic reproductive isolation, which results in the irreversible divergence of species, is commonly accompanied by hybrid sterility, necrosis/weakness, or lethality in the F1 or other offspring generations. Here we show that the loss of function of HWS1 and HWS2, a couple of duplicated paralogs, together confer complete interspecific incompatibility between Asian and African rice. Both of these non-Mendelian determinants encode the putative Esa1-associated factor 6 (EAF6) protein, which functions as a characteristic subunit of the histone H4 acetyltransferase complex regulating transcriptional activation via genome-wide histone modification. The proliferating tapetum and inappropriate polar nuclei arrangement cause defective pollen and seeds in F2 hybrid offspring due to the recombinant HWS1/2-mediated misregulation of vitamin (biotin and thiamine) metabolism and lipid synthesis. Evolutionary analysis of HWS1/2 suggests that this gene pair has undergone incomplete lineage sorting (ILS) and multiple gene duplication events during speciation. Our findings have not only uncovered a pair of speciation genes that control hybrid breakdown but also illustrate a passive mechanism that could be scaled up and used in the guidance and optimization of hybrid breeding applications for distant hybridization.

Phenotypic heterogeneity in human genetic diseases: ultrasensitivity-mediated threshold effects as a unifying molecular mechanism.

Phenotypic heterogeneity is very common in genetic systems and in human diseases and has important consequences for disease diagnosis and treatment. In addition to the many genetic and non-genetic (e.g., epigenetic, environmental) factors reported to account for part of the heterogeneity, we stress the importance of stochastic fluctuation and regulatory network topology in contributing to phenotypic heterogeneity. We argue that a threshold effect is a unifying principle to explain the phenomenon; that ultrasensitivity is the molecular mechanism for this threshold effect; and discuss the three conditions for phenotypic heterogeneity to occur. We suggest that threshold effects occur not only at the cellular level, but also at the organ level. We stress the importance of context-dependence and its relationship to pleiotropy and edgetic mutations. Based on this model, we provide practical strategies to study human genetic diseases. By understanding the network mechanism for ultrasensitivity and identifying the critical factor, we may manipulate the weak spot to gently nudge the system from an ultrasensitive state to a stable non-disease state. Our analysis provides a new insight into the prevention and treatment of genetic diseases.

Optimization of rice panicle architecture by specifically suppressing ligand-receptor pairs.

Rice panicle architecture determines the grain number per panicle and therefore impacts grain yield. The OsER1-OsMKKK10-OsMKK4-OsMPK6 pathway shapes panicle architecture by regulating cytokinin metabolism. However, the specific upstream ligands perceived by the OsER1 receptor are unknown. Here, we report that the EPIDERMAL PATTERNING FACTOR (EPF)/EPF-LIKE (EPFL) small secreted peptide family members OsEPFL6, OsEPFL7, OsEPFL8, and OsEPFL9 synergistically contribute to rice panicle morphogenesis by recognizing the OsER1 receptor and activating the mitogen-activated protein kinase cascade. Notably, OsEPFL6, OsEPFL7, OsEPFL8, and OsEPFL9 negatively regulate spikelet number per panicle, but OsEPFL8 also controls rice spikelet fertility. A osepfl6 osepfl7 osepfl9 triple mutant had significantly enhanced grain yield without affecting spikelet fertility, suggesting that specifically suppressing the OsEPFL6-OsER1, OsEPFL7-OsER1, and OsEPFL9-OsER1 ligand-receptor pairs can optimize rice panicle architecture. These findings provide a framework for fundamental understanding of the role of ligand-receptor signaling in rice panicle development and demonstrate a potential method to overcome the trade-off between spikelet number and fertility.

Long-read-based Genome Assembly of Drosophila gunungcola Reveals Fewer Chemosensory Genes in Flower-breeding Species.

Drosophila gunungcola exhibits reproductive activities on the fresh flowers of several plant species and is an emerging model to study the co-option of morphological and behavioral traits in male courtship display. Here, we report a near-chromosome-level genome assembly that was constructed based on long-read PacBio sequencing data (with ∼66× coverage) and annotated with the assistant from RNA-seq transcriptome data of whole organisms at various developmental stages. A nuclear genome of 189 Mb with 13,950 protein-coding genes and a mitogenome of 17.5 kb were acquired. Few interchromosomal rearrangements were found in the comparisons of synteny with Drosophila elegans, its sister species, and Drosophila melanogaster, suggesting that the gene compositions on each Muller element are evolutionarily conserved. Loss events of several OR and IR genes in D. gunungcola and D. elegans were revealed when orthologous genomic regions were compared across species in the D. melanogaster species group. This high-quality reference genome will facilitate further comparative studies on traits related to the evolution of sexual behavior and diet specialization.

Weak gene-gene interaction facilitates the evolution of gene expression plasticity.

BACKGROUND: Individual organisms may exhibit phenotypic plasticity when they acclimate to different conditions. Such plastic responses may facilitate or constrain the adaptation of their descendant populations to new environments, complicating their evolutionary trajectories beyond the genetic blueprint. Intriguingly, phenotypic plasticity itself can evolve in terms of its direction and magnitude during adaptation. However, we know little about what determines the evolution of phenotypic plasticity, including gene expression plasticity. Recent laboratory-based studies suggest dominance of reversing gene expression plasticity-plastic responses that move the levels of gene expression away from the new optima. Nevertheless, evidence from natural populations is still limited. RESULTS: Here, we studied gene expression plasticity and its evolution in the montane and lowland populations of an elevationally widespread songbird-the Rufous-capped Babbler (Cyanoderma ruficeps)-with reciprocal transplant experiments and transcriptomic analyses; we set common gardens at altitudes close to these populations' native ranges. We confirmed the prevalence of reversing plasticity in genes associated with altitudinal adaptation. Interestingly, we found a positive relationship between magnitude and degree of evolution in gene expression plasticity, which was pertinent to not only adaptation-associated genes but also the whole transcriptomes from multiple tissues. Furthermore, we revealed that genes with weaker expressional interactions with other genes tended to exhibit stronger plasticity and higher degree of plasticity evolution, which explains the positive magnitude-evolution relationship. CONCLUSIONS: Our experimental evidence demonstrates that species may initiate their adaptation to new habitats with genes exhibiting strong expression plasticity. We also highlight the role of expression interdependence among genes in regulating the magnitude and evolution of expression plasticity. This study illuminates how the evolution of phenotypic plasticity in gene expression facilitates the adaptation of species to challenging environments in nature.

An α/ß hydrolase family member negatively regulates salt tolerance but promotes flowering through three distinct functions in rice.

Ongoing soil salinization drastically threatens crop growth, development, and yield worldwide. It is therefore crucial that we improve salt tolerance in rice by exploiting natural genetic variation. However, many salt-responsive genes confer undesirable phenotypes and therefore cannot be effectively applied to practical agricultural production. In this study, we identified a quantitative trait locus for salt tolerance from the African rice species Oryza glaberrima and named it as Salt Tolerance and Heading Date 1 (STH1). We found that STH1 regulates fatty acid metabolic homeostasis, probably by catalyzing the hydrolytic degradation of fatty acids, which contributes to salt tolerance. Meanwhile, we demonstrated that STH1 forms a protein complex with D3 and a vital regulatory factor in salt tolerance, OsHAL3, to regulate the protein abundance of OsHAL3 via the 26S proteasome pathway. Furthermore, we revealed that STH1 also serves as a co-activator with the floral integrator gene Heading date 1 to balance the expression of the florigen gene Heading date 3a under different circumstances, thus coordinating the regulation of salt tolerance and heading date. Notably, the allele of STH1 associated with enhanced salt tolerance and high yield is found in some African rice accessions but barely in Asian cultivars. Introgression of the STH1HP46 allele from African rice into modern rice cultivars is a desirable approach for boosting grain yield under salt stress. Collectively, our discoveries not only provide conceptual advances on the mechanisms of salt tolerance and synergetic regulation between salt tolerance and flowering time but also offer potential strategies to overcome the challenges resulted from increasingly serious soil salinization that many crops are facing.

ChemistGA: A Chemical Synthesizable Accessible Molecular Generation Algorithm for Real-World Drug Discovery.

Many deep learning (DL)-based molecular generative models have been proposed to design novel molecules. These models may perform well on benchmarks, but they usually do not take real-world constraints into account, such as available training data set, synthetic accessibility, and scaffold diversity in drug discovery. In this study, a new algorithm, ChemistGA, was proposed by combining the traditional heuristic algorithm with DL, in which the crossover of the traditional genetic algorithm (GA) was redefined by DL in conjunction with GA, and an innovative backcrossing operation was implemented to generate desired molecules. Our results clearly show that ChemistGA not only retains the strength of the traditional GA but also greatly enhances the synthetic accessibility and success rate of the generated molecules with desired properties. Calculations on the two benchmarks illustrate that ChemistGA achieves impressive performance among the state-of-the-art baselines, and it opens a new avenue for the application of generative models to real-world drug discovery scenarios.

TocoDecoy: A New Approach to Design Unbiased Datasets for Training and Benchmarking Machine-Learning Scoring Functions.

Development of accurate machine-learning-based scoring functions (MLSFs) for structure-based virtual screening against a given target requires a large unbiased dataset with structurally diverse actives and decoys. However, most datasets for the development of MLSFs were designed for traditional SFs and may suffer from hidden biases and data insufficiency. Hereby, we developed a new approach named Topology-based and Conformation-based decoys generation (TocoDecoy), which integrates two strategies to generate decoys by tweaking the actives for a specific target, to generate unbiased and expandable datasets for training and benchmarking MLSFs. For hidden bias evaluation, the performance of InteractionGraphNet (IGN) trained on the TocoDecoy, LIT-PCBA, and DUD-E-like datasets was assessed. The results illustrate that the IGN model trained on the TocoDecoy dataset is competitive with that trained on the LIT-PCBA dataset but remarkably outperforms that trained on the DUD-E dataset, suggesting that the decoys in TocoDecoy are unbiased for training and benchmarking MLSFs.

A genetic module at one locus in rice protects chloroplasts to enhance thermotolerance.

How the plasma membrane senses external heat-stress signals to communicate with chloroplasts to orchestrate thermotolerance remains elusive. We identified a quantitative trait locus, Thermo-tolerance 3 (TT3), consisting of two genes, TT3.1 and TT3.2, that interact together to enhance rice thermotolerance and reduce grain-yield losses caused by heat stress. Upon heat stress, plasma membrane-localized E3 ligase TT3.1 translocates to the endosomes, on which TT3.1 ubiquitinates chloroplast precursor protein TT3.2 for vacuolar degradation, implying that TT3.1 might serve as a potential thermosensor. Lesser accumulated, mature TT3.2 proteins in chloroplasts are essential for protecting thylakoids from heat stress. Our findings not only reveal a TT3.1-TT3.2 genetic module at one locus that transduces heat signals from plasma membrane to chloroplasts but also provide the strategy for breeding highly thermotolerant crops.

Epigenomic signatures on paralogous genes reveal underappreciated universality of active histone codes adopted across animals.

The results of conventional gene-based analyses which combine epigenome and transcriptome data, including those conducted by the ENCODE/modENCODE projects, suggest various histone modifications performing regulatory functions in controlling mRNA expression (referred to as a histone code) in several model animals. While some histone codes were found to be universally adopted across organisms, "species-specific" histone codes have also been defined. We found that the characterization of these histone codes was confounded by factors (e.g. gene essentiality, expression breadth) that are independent of, but correlated with, gene expression levels. Hence, we attempted to decode histone marks in mouse (Mus musculus), fly (Drosophila melanogaster), and worm (Caenorhabditis elegans) genomes by examining ratios of RNA sequencing (and chromatin immunoprecipitation sequencing) intensities between paralog genes to remove confounding effects that would otherwise be present in a gene-based approach. With this paralog-based approach, associations between four histone modifications (H3K4me3, H3K27ac, H3K9ac, and H3K36me3) and gene expression are substantially revised. For example, we demonstrate that H3K27ac and H3K9ac represent universal active marks in promoters, rather than worm-specific marks as previously reported. Second, acting regions of the studied active marks that are common across species (and across a wide range of tissues at different developmental stages) were found to extend beyond the previously defined regions. Thus, it appears that the active histone codes analyzed have a universality that has previously been underappreciated. Our results suggested that these universal codes, including those previously considered species-specific, could have an ancient origin, and are important in regulating animal gene expression abundance.

InteractionGraphNet: A Novel and Efficient Deep Graph Representation Learning Framework for Accurate Protein-Ligand Interaction Predictions.

Accurate quantification of protein-ligand interactions remains a key challenge to structure-based drug design. However, traditional machine learning (ML)-based methods based on handcrafted descriptors, one-dimensional protein sequences, and/or two-dimensional graph representations limit their capability to learn the generalized molecular interactions in 3D space. Here, we proposed a novel deep graph representation learning framework named InteractionGraphNet (IGN) to learn the protein-ligand interactions from the 3D structures of protein-ligand complexes. In IGN, two independent graph convolution modules were stacked to sequentially learn the intramolecular and intermolecular interactions, and the learned intermolecular interactions can be efficiently used for subsequent tasks. Extensive binding affinity prediction, large-scale structure-based virtual screening, and pose prediction experiments demonstrated that IGN achieved better or competitive performance against other state-of-the-art ML-based baselines and docking programs. More importantly, such state-of-the-art performance was proven from the successful learning of the key features in protein-ligand interactions instead of just memorizing certain biased patterns from data.

A rice QTL GS3.1 regulates grain size through metabolic-flux distribution between flavonoid and lignin metabolons without affecting stress tolerance.

Grain size is a key component trait of grain weight and yield. Numbers of quantitative trait loci (QTLs) have been identified in various bioprocesses, but there is still little known about how metabolism-related QTLs influence grain size and yield. The current study report GS3.1, a QTL that regulates rice grain size via metabolic flux allocation between two branches of phenylpropanoid metabolism. GS3.1 encodes a MATE (multidrug and toxic compounds extrusion) transporter that regulates grain size by directing the transport of p-coumaric acid from the p-coumaric acid biosynthetic metabolon to the flavonoid biosynthetic metabolon. A natural allele of GS3.1 was identified from an African rice with enlarged grains, reduced flavonoid content and increased lignin content in the panicles. Notably, the natural allele of GS3.1 caused no alterations in other tissues and did not affect stress tolerance, revealing an ideal candidate for breeding efforts. This study uncovers insights into the regulation of grain size though metabolic-flux distribution. In this way, it supports a strategy of enhancing crop yield without introducing deleterious side effects on stress tolerance mechanisms.

Hyperbolic relational graph convolution networks plus: a simple but highly efficient QSAR-modeling method.

Accurate predictions of druggability and bioactivities of compounds are desirable to reduce the high cost and time of drug discovery. After more than five decades of continuing developments, quantitative structure-activity relationship (QSAR) methods have been established as indispensable tools that facilitate fast, reliable and affordable assessments of physicochemical and biological properties of compounds in drug-discovery programs. Currently, there are mainly two types of QSAR methods, descriptor-based methods and graph-based methods. The former is developed based on predefined molecular descriptors, whereas the latter is developed based on simple atomic and bond information. In this study, we presented a simple but highly efficient modeling method by combining molecular graphs and molecular descriptors as the input of a modified graph neural network, called hyperbolic relational graph convolution network plus (HRGCN+). The evaluation results show that HRGCN+ achieves state-of-the-art performance on 11 drug-discovery-related datasets. We also explored the impact of the addition of traditional molecular descriptors on the predictions of graph-based methods, and found that the addition of molecular descriptors can indeed boost the predictive power of graph-based methods. The results also highlight the strong anti-noise capability of our method. In addition, our method provides a way to interpret models at both the atom and descriptor levels, which can help medicinal chemists extract hidden information from complex datasets. We also offer an HRGCN+'s online prediction service at https://quantum.tencent.com/hrgcn/.

Could graph neural networks learn better molecular representation for drug discovery? A comparison study of descriptor-based and graph-based models.

Graph neural networks (GNN) has been considered as an attractive modelling method for molecular property prediction, and numerous studies have shown that GNN could yield more promising results than traditional descriptor-based methods. In this study, based on 11 public datasets covering various property endpoints, the predictive capacity and computational efficiency of the prediction models developed by eight machine learning (ML) algorithms, including four descriptor-based models (SVM, XGBoost, RF and DNN) and four graph-based models (GCN, GAT, MPNN and Attentive FP), were extensively tested and compared. The results demonstrate that on average the descriptor-based models outperform the graph-based models in terms of prediction accuracy and computational efficiency. SVM generally achieves the best predictions for the regression tasks. Both RF and XGBoost can achieve reliable predictions for the classification tasks, and some of the graph-based models, such as Attentive FP and GCN, can yield outstanding performance for a fraction of larger or multi-task datasets. In terms of computational cost, XGBoost and RF are the two most efficient algorithms and only need a few seconds to train a model even for a large dataset. The model interpretations by the SHAP method can effectively explore the established domain knowledge for the descriptor-based models. Finally, we explored use of these models for virtual screening (VS) towards HIV and demonstrated that different ML algorithms offer diverse VS profiles. All in all, we believe that the off-the-shelf descriptor-based models still can be directly employed to accurately predict various chemical endpoints with excellent computability and interpretability.

Characterization of TMAO productivity from carnitine challenge facilitates personalized nutrition and microbiome signatures discovery.

The capability of gut microbiota in degrading foods and drugs administered orally can result in diversified efficacies and toxicity interpersonally and cause significant impact on human health. Production of atherogenic trimethylamine N-oxide (TMAO) from carnitine is a gut microbiota-directed pathway and varies widely among individuals. Here, we demonstrated a personalized TMAO formation and carnitine bioavailability from carnitine supplements by differentiating individual TMAO productivities with a recently developed oral carnitine challenge test (OCCT). By exploring gut microbiome in subjects characterized by TMAO producer phenotypes, we identified 39 operational taxonomy units that were highly correlated to TMAO productivity, including Emergencia timonensis, which has been recently discovered to convert γ-butyrobetaine to TMA in vitro. A microbiome-based random forest classifier was therefore constructed to predict the TMAO producer phenotype (AUROC = 0.81) which was then validated with an external cohort (AUROC = 0.80). A novel bacterium called Ihubacter massiliensis was also discovered to be a key microbe for TMA/TMAO production by using an OCCT-based humanized gnotobiotic mice model. Simply combining the presence of E. timonensis and I. massiliensis could account for 43% of high TMAO producers with 97% specificity. Collectively, this human gut microbiota phenotype-directed approach offers potential for developing precision medicine and provides insights into translational research. Video Abstract.

Comparative Transcriptomics across Nematode Life Cycles Reveal Gene Expression Conservation and Correlated Evolution in Adjacent Developmental Stages.

Nematodes are highly abundant animals with diverse habitats and lifestyles. Some are free living whereas others parasitize animals or plants, and among the latter, infection abilities change across developmental stages to infect hosts and complete life cycles. To determine the relationship between transcriptome evolution and morphological divergences among nematodes, we compared 48 transcriptomes of different developmental stages across eight nematode species. The transcriptomes were clustered broadly into embryo, larva, and adult stages, with the developmental plastic stages were separated from common larval stages within the larval branch. This suggests that development was the major determining factor after lifestyle changes, such as parasitism, during transcriptome evolution. Such patterns were partly accounted for by tissue-specific genes-such as those in oocytes and the hypodermis-being expressed at different proportions. Although nematodes typically have 3-5 larval stages, the transcriptomes for these stages were found to be highly correlated within each species, suggesting high similarity among larval stages across species. For the Caenorhabditis elegans-Caenorhabditis briggsae and Strongyloides stercoralis-Strongyloides venezuelensis comparisons, we found that ∼50% of genes were expressed at multiple stages, whereas half of their orthologs were also expressed in multiple but different stages. Such frequent changes in expression have resulted in concerted transcriptome evolution across adjacent stages, thus generating species-specific transcriptomes over the course of nematode evolution. Our study provides a first insight into the evolution of nematode transcriptomes beyond embryonic development.