Phenotype Prediction Research Articles

Comprehensive genomic analysis and selection signature detection in endangered Beigang pigs using whole-genome sequencing data.

The Beigang pig was recently identified as one of the endangered breeds during a Chinese indigenous pig genetic resource survey. The Beigang breed is notable for its remarkable roughage tolerance and high reproductive capacity according to historical records. Morphologically, the Beigang pig resembles many indigenous pigs in eastern China, especially in its large ears. This makes the Beigang pig a valuable reference for studying the genetic mechanisms on large ear size in pigs. However, there is currently a lack of clear understanding regarding the genetic structure and inbreeding levels of the Beigang pig population. This study used whole-genome sequencing data from Beigang pig (N = 145 pigs) and integrated genetic information from commercial pigs and indigenous pigs in eastern China to conduct a comprehensive analysis of the Beigang pig's genetic structure. Three selection signal detection methods-runs of homozygosity, fixation index, and integrated haplotype score-were employed to explore the differences in genomic selection signatures between Beigang pig and other pig populations. Additionally, we used a public project for regulatory variants discovery and molecular phenotype prediction in farm animal species called FarmGtex to explore the expression of three genes (WIF1, LEMD3, and MSRB3) related to ear size in Beigang pig. This research identified five homozygous variant sites in the WIF1 gene as important candidate loci potentially influencing ear size in Beigang pig. The results indicate that the Beigang pig holds a unique status among Chinese indigenous pigs, characterized by high genetic diversity and low levels of inbreeding. The study also revealed that WIF1 may play a significant role in influencing ear size in this breed. These findings contribute to a deeper understanding of the population structure and genetic characteristics of Beigang pig.

Clinical considerations on antimicrobial resistance potential of complex microbiological samples

Antimicrobial resistance (AMR) is one of our greatest public health challenges. Targeted use of antibiotics (ABs) can reduce the occurrence and spread of AMR and boost the effectiveness of treatment. This requires knowledge of the AB susceptibility of the pathogens involved in the disease. Therapeutic recommendations based on classical AB susceptibility testing (AST) are based on the analysis of only a fraction of the bacteria present in the disease process. Next and third generation sequencing technologies allow the identification of antimicrobial resistance genes (ARGs) present in a bacterial community. Using this metagenomic approach, we can map the antimicrobial resistance potential (AMRP) of a complex, multi-bacterial microbial sample. To understand the interpretiveness of AMRP, the concordance between phenotypic AMR properties and ARGs was investigated by analyzing data from 574 Escherichia coli strains of five different studies. The overall results show that for 44% of the studied ABs, phenotypically resistant strains are genotypically associated with a 90% probability of resistance, while for 92% of the ABs, the phenotypically susceptible strains are genotypically susceptible with a 90% probability. ARG detection showed a phenotypic prediction with at least 90% confidence in 67% of ABs. The probability of detecting a phenotypically susceptible strain as resistant based on genotype is below 5% for 92% of ABs. While the probability of detecting a phenotypically resistant strain as susceptible based on genotype is below 5% for 44% of ABs. We can assume that these strain-by-strain concordance results are also true for bacteria in complex microbial samples, and conclude that AMRP obtained from metagenomic ARG analysis can help choose efficient ABs. This is illustrated using AMRP by a canine external otitis sample.

A Behavioral Screening Method for Predicting PTSD-like Phenotypes: Novel Application to Female Rats.

Only a small percentage of trauma-exposed subjects develop PTSD, with females being twice as likely. Most rodent models focus on males and fail to account for inter-individual variability in females. We tested a behavioral PTSD model in female rats to distinguish between susceptible and resilient individuals. In Experiment 1, female rats underwent footshocks paired with social isolation, a PTSD risk factor. They were re-exposed to the conditioned context to test memory retention, and assessed in the Elevated Plus Maze (EPM) and Social Interaction (SI) tests for anxiety and social behavior. Footshock-exposed rats showed fear memory retention up to 16 days, indicated by elevated freezing behavior during re-exposure. They also exhibited reduced exploration in the EPM and less SI time compared to controls. In Experiment 2, we classified rats into normal responders, susceptible, and resilient groups based on locomotor activity after trauma, correlating with memory retention and anxiety. Unlike existing models focused on males and lacking predictive variables before trauma, our method identifies PTSD-like susceptibility and resilience in female rats by using exploratory behavior as a predictor before trauma exposure. Exploratory activity in a novel environment after trauma and before extinction is a reliable predictor of PTSD-like phenotypes and differentiates between susceptible and resilient female rats.

Metabolic modeling identifies determinants of thermal growth responses in Arabidopsis thaliana.

Temperature is a critical environmental factor affecting nearly all plant processes, including growth, development, and yield. Yet, despite decades of research, we lack the ability to predict plant performance at different temperatures, limiting the development of climate-resilient crops. Further, there is a pressing need to bridge the gap between the prediction of physiological and molecular traits to improve our understanding and manipulation of plant temperature responses. Here, we developed the first enzyme-constrained model of Arabidopsis thaliana's metabolism, facilitating predictions of growth-related phenotypes at different temperatures. We showed that the model can be employed for in silico identification of genes that affect plant growth at suboptimal growth temperature. Using mutant lines, we validated the genes predicted to affect plant growth, demonstrating the potential of metabolic modeling in accurately predicting plant thermal responses. The temperature-dependent enzyme-constrained metabolic model provides a template that can be used for developing sophisticated strategies to engineer climate-resilient crops.

Hypervirulent Klebsiella pneumoniae have better clinical outcomes than classical Klebsiella pneumoniae for lower respiratory tract infection patients

BackgroundThe clinical outcomes and microbiological features of lower respiratory tract infections (LRTIs) caused by hypervirulent Klebsiella pneumoniae (hvKp) and classical Klebsiella pneumoniae (cKp) have not been well understood.MethodsThis study collected 287 non-repetitive Klebsiella pneumoniae isolates from 287 LRTI patients. All these strains underwent annotation for resistance and virulence factors, with 141 strains undergoing mouse infection experiments to assess their virulence. The primary clinical outcomes of these patients were evaluated, including intensive care unit (ICU) admission and in-hospital mortality rates.ResultsA total of 46 capsule serotypes were identified. Among these isolates subjected to mouse infection experiments, the proportions of strains exhibiting hypervirulent phenotypes were 92.6% (25/27), 92.1% (35/38), 80% (4/5), 25% (1/4), 10.5% (2/19), and 7.1% (1/14) for K2, K1, K20, K54, K47, and K25, respectively. Therefore, K1, K2, and K20 K. pneumoniae were defined as hvKp. In addition, the rates of ICU admission and in-hospital mortality for hvKp-infected patients were significantly lower than those of cKp-infected patients (51.4% vs. 65.9%, χ2 = 4.722, p = 0.03 and 8.6% vs. 29%, χ2 = 12.133, p < 0.001). Notably, among the cKp group, the cKp-ST11 subgroup had higher rates of ICU admission (77% vs. 58.5%, χ2 = 7.981, p = 0.005) and in-hospital mortality (44.8% vs. 18.5%, χ2 = 17.585, p < 0.001) than cKp-nonST11 subgroup.ConclusionsThese findings suggest that capsule serotype is a more accurate factor for the prediction of the virulence phenotype, while hvKp have better clinical outcomes than cKp for LRTI patients. Furthermore, the cKp-ST11 subgroup has the worst prognosis than cKp-nonST11 subgroup.

Molecular allergy diagnosis enabling personalized medicine.

Allergic patients are characterized by complex and patient-specific IgE sensitization profiles to various allergens, which are accompanied by different phenotypes of allergic disease. Molecular allergy (MA) diagnosis establishes the patient's IgE reactivity profile at a molecular allergen level and has moved allergology into the "Precision Medicine" era. Molecular allergology started in the late 1980s with the isolation of the first allergen-encoding DNA sequences. Already in 2002, the first allergen microarrays were developed for the assessment of complex IgE sensitization patterns. Recombinant allergens are used for a precise definition of personal IgE reactivity profiles, identification of genuine IgE sensitization to allergen sources for refined prescription of allergen-specific immunotherapy and allergen avoidance diagnosis of co- versus cross-sensitization, epidemiological studies and prediction of symptoms, phenotypes, and development allergic disease. For example, molecular IgE sensitization patterns associated with more severe respiratory allergies, severe food allergy and allergy to honeybee or vespids are already established. The implementation of MA diagnosis into daily clinical practice requires continuous medical education, training of doctors in MA diagnosis and may be facilitated by clinical decision support systems such as diagnostic algorithms which may take advantage of artificial intelligence (AI).

Associations between IL-33 gene and IL1RL1 gene variants in periodontitis

Background: Periodontitis is initiated by a dysbiosis in the subgingival microbial biofilm and can be related to host genetic factors. This study investigated association between periodontitis and single nucleotide variants (SNVs) in the IL-33 and IL1RL1 genes. Methods: This cross-sectional study involved 359 individuals from a public health service in Brazil. Structured questionnaire was used to collect health status and socioeconomic, demographic and behavioral characteristics. Periodontitis was diagnosed by clinical periodontal examination. Subgingival biofilm was collected at the deepest site of each sextant, and biofilm bacterial DNA was amplified by real time polymerase chain reaction (qPCR) to determine relative quantification of pathogens. Peripheral blood was collected for genomic DNA extraction and SNV genotyping was performed by qPCR. Logistic regression model was used to obtain association measures (95% confidence interval), by ussing the additive model. Results: The C allele variant of IL33 (rs2381416) was inversely associated with periodontitis, even after adjusting for the confounding covariates (p &lt; 0.01 ORadjusted: 0.45; CI: 0.24-0.84), and with the presence of the Aggregatibacter actinomycetemcomitans (Aa) pathogen, adjusting for the same covariates (p &lt; 0.01; ORadjusted: 0.46; CI: 0.27-0.76). Inverse association between this SNV and periodontitis was observed (p = 0.02; ORadjusted: 0.46; CI: 0.28-0.76) using additive genotypic model. Conclusions: Frequency of C allele variant of IL-33 (rs2381416) was lower in individuals with periodontitis and in individuals with relatively higher levels of Aa. Investigations of this variant as a potential predictor of the protective phenotype in the context of periodontitis are needed. This study will contribute to the training of health professionals involved in the treatment of periodontitis.

Genotype–Phenotype Correlation in a Large Cohort of Eastern Sicilian Patients Affected by Phenylketonuria: Newborn Screening Program, Clinical Features, and Follow-Up

Background: Phenylketonuria (PKU) is an autosomal recessive disorder caused by mutations in the phenylalanine hydroxylase (PAH) gene, leading to impaired amino acid metabolism. Early diagnosis through newborn screening (NBS) enables prompt treatment, preventing neurological complications. This study aims to describe the genetic and phenotypic spectrum of PKU and mild hyperphenylalaninemia (m-HPA) in patients diagnosed at the Department of Inborn Errors of Metabolism and Newborn Screening, Hospital G. Rodolico-S. Marco, Catania, over four decades (1987–2023). Materials and Methods: The retrospective analysis included 102 patients with elevated blood phenylalanine (Phe) levels born in Sicily and followed at the Institute. The phenotype evaluation comprised the Phe levels at birth/diagnosis, dietary tolerance, and sapropterin dihydrochloride responsiveness. The dietary compliance and Phe/Tyr ratios were assessed and compared across phenotypic classes and age groups. Results: Of 102 patients, 34 were classified as having classic PKU, 9 as having moderate PKU, 26 as having mild PKU, and 33 as having m-HPA, with a median age of 21.72 years. Common PAH variants included c.1066-11G&gt;A (26/204 alleles), c.782G&gt;A (18/204 alleles), and c.165delT (13/204 alleles). The phenotypes sometimes diverged from the genotype predictions, emphasizing dietary tolerance over the initial Phe levels for classification: m-HPA was statistically associated with a higher dietary tolerance (p &lt; 0.001) compared to the classic, moderate, or mild forms of PKU. Conclusions: This study highlights the importance of large databases (e.g., BioPKU) for phenotype prediction and treatment optimization. Regular assessment of Phe/Tyr ratios is crucial for monitoring adherence and health. Phenotype determination, dietary management, and emerging therapies (Pegvaliase and gene therapy) are key to improving outcomes for PKU patients.

G2PDeep-v2: a web-based deep-learning framework for phenotype prediction and biomarker discovery for all organisms using multi-omics data.

The G2PDeep-v2 server is a web-based platform powered by deep learning, for phenotype prediction and markers discovery from multi-omics data in any organisms including humans, plants, animals, and viruses. The server provides multiple services for researchers to create deep-learning models through an interactive interface and train these models using an automated hyperparameter tuning algorithm on high-performance computing resources. Users can visualize the results of phenotype and markers predictions and perform Gene Set Enrichment Analysis for the significant markers to provide insights into the molecular mechanisms underlying complex diseases, conditions and other biological phenotypes being studied. The G2PDeep-v2 server is publicly available at https://g2pdeep.org/ and can be utilized for all organisms.

CAGI6 ID panel challenge: assessment of phenotype and variant predictions in 415 children with neurodevelopmental disorders (NDDs).

The Genetics of Neurodevelopmental Disorders Lab in Padua provided a new intellectual disability (ID) Panel challenge for computational methods to predict patient phenotypes and their causal variants in the context of the Critical Assessment of the Genome Interpretation, 6th edition (CAGI6). Eight research teams submitted a total of 30 models to predict phenotypes based on the sequences of 74 genes (VCF format) in 415 pediatric patients affected by Neurodevelopmental Disorders (NDDs). NDDs are clinically and genetically heterogeneous conditions, with onset in infant age. Here, we assess the ability and accuracy of computational methods to predict comorbid phenotypes based on clinical features described in each patient and their causal variants. We also evaluated predictions for possible genetic causes in patients without a clear genetic diagnosis. Like the previous ID Panel challenge in CAGI5, seven clinical features (ID, ASD, ataxia, epilepsy, microcephaly, macrocephaly, hypotonia), and variants (Pathogenic/Likely Pathogenic, Variants of Uncertain Significance and Risk Factors) were provided. The phenotypic traits and variant data of 150 patients from the CAGI5 ID Panel Challenge were provided as training set for predictors. The CAGI6 challenge confirms CAGI5 results that predicting phenotypes from gene panel data is highly challenging, with AUC values close to random, and no method able to predict relevant variants with both high accuracy and precision. However, a significant improvement is noted for the best method, with recall increasing from 66% to 82%. Several groups also successfully predicted difficult-to-detect variants, emphasizing the importance of variants initially excluded by the Padua NDD Lab.

Phenotypic antibiotic resistance prediction using antibiotic resistance genes and machine learning models in Mannheimia haemolytica.

Mannheimia haemolytica is one of the most common causative agents of bovine respiratory disease (BRD); however, antibiotic resistance in this species is increasing, making treatment more difficult. Integrative-conjugative elements (ICE), a subset of mobile genetic elements (MGE), encoding up to 100 genes have been reported in Mannheimia haemolytica genomes to confer multidrug resistance, including resistance to antibiotics commonly used in the treatment of BRD. However, the presence of antibiotic resistance genes (ARGs) does not always agree with phenotypic resistance. Prior investigations have reported an overall phenotype-genotype concordance less than 75 % in BRD pathogens. The objective of the current study was to compare genotype-phenotype concordance either by annotating known resistance genes in genomes or predicting antibiotic resistance determinants de novo with machine learning (ML). The genotype-phenotype concordance rates of ARGs were generally > 90 %, while that of ML models were > 80 %. For all seven antibiotics (danofloxacin, enrofloxacin, florfenicol, tetracycline, tildipirosin, tilmicosin, and tulathromycin), the genotype-phenotype concordance rates were more accurate with ARGs. The annotations of ML models for all antibiotics included various types of sequences, including coding sequences such as DNA topoisomerase IV (danofloxacin) and non-coding sequences near tetracycline genes (multiple antibiotics), MGE (tetracycline and tildipirosin), or virulence genes (danofloxacin and enrofloxacin). When tested on an external set of isolates for validation, the best predictor of antibiotic resistance performed similarly to the training/testing datasets for each antibiotic. Incorporating single nucleotide polymorphisms and ARGs unknown during previous studies resulted in higher concordance rates (>90 %) for fluoroquinolones and tilmicosin, respectively. By finding increased concordance rates for known ARGs, this study was able to show that ARGs should continue to be utilized to predict phenotypic resistance.

Reconsidering remission in recurrent late-life depression: clinical presentation and phenotypic predictors of relapse following successful antidepressant treatment.

Late-life depression (LLD) is characterized by repeated recurrent depressive episodes even with maintenance treatment. It is unclear what clinical and cognitive phenotypic characteristics present during remission predict future recurrence. Participants (135 with remitted LLD and 69 comparison subjects across three institutions) completed baseline phenotyping, including psychiatric, medical, and social history, psychiatric symptom and personality trait assessment, and neuropsychological testing. Participants were clinically assessed every two months for two years while receiving standard antidepressant treatment. Analyses examined group differences in phenotypic measure using general linear models. Concurrent associations between phenotypic measures and diagnostic groups were examined using LASSO logistic regression. Sixty (44%) LLD participants experienced a relapse over the two-year period. Numerous phenotypic measures across all domains differed between remitted LLD and comparison participants. Only residual depressive symptom severity, rumination, medical comorbidity, and executive dysfunction significantly predicted LLD classification. Fewer measures differed between relapsing and sustained remission LLD subgroups, with the relapsing group exhibiting greater antidepressant treatment intensity, greater fatigue, rumination, and disability, higher systolic blood pressure, greater life stress and lower instrumental social support. Relapsing group classification was informed by antidepressant treatment intensity, lower instrumental social support, and greater life stress. A wide range of phenotypic factors differed between remitted LLD and comparison groups. Fewer measures differed between relapsing and sustained remission LLD subgroups, with less social support and greater stress informing vulnerability to subsequent relapse. This research suggests potential targets for relapse prevention and emphasizes the need for clinically translatable relapse biomarkers to inform care.

Improving polygenic prediction from summary data by learning patterns of effect sharing across multiple phenotypes.

Polygenic prediction of complex trait phenotypes has become important in human genetics, especially in the context of precision medicine. Recently, mr.mash, a flexible and computationally efficient method that models multiple phenotypes jointly and leverages sharing of effects across such phenotypes to improve prediction accuracy, was introduced. However, a drawback of mr.mash is that it requires individual-level data, which are often not publicly available. In this work, we introduce mr.mash-rss, an extension of the mr.mash model that requires only summary statistics from Genome-Wide Association Studies (GWAS) and linkage disequilibrium (LD) estimates from a reference panel. By using summary data, we achieve the twin goal of increasing the applicability of the mr.mash model to data sets that are not publicly available and making it scalable to biobank-size data. Through simulations, we show that mr.mash-rss is competitive with, and often outperforms, current state-of-the-art methods for single- and multi-phenotype polygenic prediction in a variety of scenarios that differ in the pattern of effect sharing across phenotypes, the number of phenotypes, the number of causal variants, and the genomic heritability. We also present a real data analysis of 16 blood cell phenotypes in the UK Biobank, showing that mr.mash-rss achieves higher prediction accuracy than competing methods for the majority of traits, especially when the data set has smaller sample size.

GEFormer: a Genomic Prediction Method of Genotype-Environment Interaction in Maize by Integrating Gating Mechanism MLP and Linear Attention Mechanism.

The integration of genotypic and environmental data can enhance the prediction accuracy of field traits of crops. The existing genomic prediction methods fail to consider the environmental factors and do not consider the real growing environment of crops, resulting in low genomic prediction accuracy. In this work, we propose a genotype-environment interaction genomic prediction method in maize, called GEFormer, based on integrating the gating mechanism MLP and linear attention mechanism. Firstly, it uses gated multilayer perceptron (gMLP) to extract the local and global features among SNPs. Then, the Omni-dimensional Dynamic Convolution is used to extract the dynamic and comprehensive features of multiple environmental factors within each day in the consideration of the real growth pattern of crops. The linear attention mechanism is used to capture the temporal features of environmental changes. Finally, it uses the gating mechanism to fuse the genomic and environmental features effectively. We validate the accuracy of GEFormer in predicting important agronomic traits of maize, rice and wheat in three experimental scenarios: untested genotypes in tested environments, tested genotypes in untested environments, untested genotypes in untested environments. Experimental results show that GEFormer outperforms six cutting-edge statistical learning methods and four machine learning methods. Furthermore, it shows great advantages in the experimental scenario of untested genotypes in untested environments. In addition, we used GEFormer into three real-world breeding applications: phenotype prediction in unknown environments, hybrid phenotype prediction using inbred population, and cross-population phenotype prediction. The results illustrate that GEFormer exhibiting better prediction performance in actual breeding scenarios, and it can be utilized to assist crop breeding.

Unlocking the diagnostic potential of electrocardiograms through information transfer from cardiac magnetic resonance imaging.

Cardiovascular diseases (CVD) can be diagnosed using various diagnostic modalities. The electrocardiogram (ECG) is a cost-effective and widely available diagnostic aid that provides functional information of the heart. However, its ability to classify and spatially localise CVD is limited. In contrast, cardiac magnetic resonance (CMR) imaging provides detailed structural information of the heart and thus enables evidence-based diagnosis of CVD, but long scan times and high costs limit its use in clinical routine. In this work, we present a deep learning strategy for cost-effective and comprehensive cardiac screening solely from ECG. Our approach combines multimodal contrastive learning with masked data modelling to transfer domain-specific information from CMR imaging to ECG representations. In extensive experiments using data from 40,044 UK Biobank subjects, we demonstrate the utility and generalisability of our method for subject-specific risk prediction of CVD and the prediction of cardiac phenotypes using only ECG data. Specifically, our novel multimodal pre-training paradigm improves performance by up to 12.19% for risk prediction and 27.59% for phenotype prediction. In a qualitative analysis, we demonstrate that our learned ECG representations incorporate information from CMR image regions of interest. Our entire pipeline is publicly available at https://github.com/oetu/MMCL-ECG-CMR.

Unveiling the ghost: machine learning's impact on the landscape of virology.

The complexity and speed of evolution in viruses with RNA genomes makes predictive identification of variants with epidemic or pandemic potential challenging. In recent years, machine learning has become an increasingly capable technology for addressing this challenge, as advances in methods and computational power have dramatically improved the performance of models and led to their widespread adoption across industries and disciplines. Nascent applications of machine learning technology to virus research have now expanded, providing new tools for handling large-scale datasets and leading to a reshaping of existing workflows for phenotype prediction, phylogenetic analysis, drug discovery and more. This review explores how machine learning has been applied to and has impacted the study of viruses, before addressing the strengths and limitations of its techniques and finally highlighting the next steps that are needed for the technology to reach its full potential in this challenging and ever-relevant research area.

A data-integrative modeling approach accurately characterizes the effects of mutations on Arabidopsis lipid metabolism.

Collections of insertional mutants have been instrumental for characterizing the functional relevance of genes in different model organisms, including Arabidopsis (Arabidopsis thaliana). However, mutations may often result in subtle phenotypes, rendering it difficult to pinpoint the function of a knocked-out gene. Here, we present a data-integrative modeling approach that enables predicting the effects of mutations on metabolic traits and plant growth. To test the approach, we gathered lipidomics data and physiological read-outs for a set of 64 Arabidopsis lines with mutations in lipid metabolism. Use of flux sums as a proxy for metabolite concentrations allowed us to integrate the relative abundance of lipids and facilitated accurate predictions of growth and biochemical phenotype in approximately 73% and 76% of the mutants, respectively, for which phenotypic data were available. Likewise, we showed that this approach can pinpoint alterations in metabolic pathways related to silent mutations. Therefore, our study paves the way for coupling model-driven characterization of mutant lines from different mutagenesis approaches with metabolomic technologies, as well as for validating knowledge structured in large-scale metabolic networks of plants and other species.

Toward generalizable phenotype prediction from single-cell morphology representations

BackgroundFunctional cell processes (e.g., molecular signaling, response to stimuli, mitosis, etc.) impact cell phenotypes, which scientists can measure with cell morphology. However, linking these measurements with phenotypes remains challenging because it requires manually annotated labels. We propose that nuclear morphology can be a predictive marker for cell phenotypes that are generalizable across contexts.MethodsWe reanalyzed a pre-labeled, publicly-available nucleus microscopy dataset from the MitoCheck consortium. We extracted single-cell morphology features using CellProfiler and DeepProfiler, which provide robust processing pipelines. We trained multinomial, multi-class elastic-net logistic regression models to classify nuclei into one of 15 phenotypes such as ‘Anaphase,’ ‘Apoptosis’, and ‘Binuclear’. We rigorously assessed performance using F1 scores, precision-recall curves, and a leave-one-image-out (LOIO) cross-validation analysis. In LOIO, we retrained models using cells from every image except one and predicted phenotype in the held-out image, repeating this procedure for all images. We evaluated each morphology feature space, a concatenated feature space, and several feature space subsets (e.g., nuclei AreaShape features only). We applied models to the Joint Undertaking in Morphological Profiling (JUMP) data to assess performance using a different dataset.ResultsIn a held-out test set, we observed an overall F1 score of 0.84. Individual phenotype scores ranged from 0.64 (moderate performance) to 0.99 (high performance). Phenotypes such as ‘Elongated’, ‘Metaphase’, and ‘Apoptosis’ showed high performance. While CellProfiler and DeepProfiler features were generally equally effective, concatenation yielded the best results for 9/15 phenotypes. LOIO showed a performance decline, indicating our model could not reliably predict phenotypes in new images. Poor performance was unrelated to illumination correction or model selection. Applied to the JUMP data, models trained using nuclear AreaShape features only increased alignment with the annotated MitoCheck data (based on UMAP space). This approach implicated many chemical and genetic perturbations known to be associated with specific phenotypes.DiscussionPoor LOIO performance demonstrates challenges of single-cell phenotype prediction in new datasets. We propose several strategies that could pave the way for more generalizable methods in single-cell phenotype prediction, which is a step toward morphology representation ontologies that would aid in cross-dataset interpretability.

Multi-view BLUP: a promising solution for post-omics data integrative prediction

Phenotypic prediction is a promising strategy for accelerating plant breeding. Data from multiple sources (called multi-view data) can provide complementary information to characterize a biological object from various aspects. By integrating multi-view information into phenotypic prediction, a multi-view best linear unbiased prediction (MVBLUP) method was proposed in this paper. To measure the importance of multiple data views, the differential evolution algorithm with an early stopping mechanism was used, by which we obtained a multi-view kinship matrix and then incorporated it into the BLUP model for phenotypic prediction. To further illustrate the characteristics of MVBLUP, we performed the empirical experiments on four multi-view datasets in different crops. Compared to the single-view method, the prediction accuracy of the MVBLUP method has improved by 0.038 to 0.201 on average. The results demonstrate that the MVBLUP is an effective integrative prediction method for multi-view data.

Current Status and Applications of Genome‐Scale Metabolic Models of Oleaginous Microorganisms

ABSTRACTOleaginous microorganisms have the unique ability to accumulate lipids that can exceed 20% of their dry cell weight under certain conditions. Despite their potential for efficient lipid production, the metabolic pathways involved are not yet fully understood, largely due to the complexity of intracellular processes and the challenges in phenotypic prediction. This review synthesizes the latest research on the application of Genome‐scale Metabolic Network Models (GSMMs) to study oleaginous microorganisms, including bacteria, cyanobacteria, yeast, microalgae, and fungi, and provides a comprehensive analysis of how GSMMs have been utilized to decipher the metabolic mechanisms behind lipid accumulation and to identify key genes involved in lipid synthesis. The review highlights the role of GSMMs in predicting cellular behavior, optimizing metabolic engineering strategies, and discusses the future directions and potential of GSMMs in enhancing lipid production in microorganisms. This comprehensive overview not only summarizes the current state of research but also identifies gaps and opportunities for further investigation in the field.