Integrating gene expression and GO classification for PCA by preclustering

Rational selection of morphological phenotypic traits to extract essential similarities in chemical perturbation in the ergosterol pathway

Terbinafine, fluconazole, and amorolfine inhibit fungal ergosterol synthesis by acting on their target enzymes at different steps in the synthetic pathway, causing the accumulation of various intermediates. We found that the effects of these three in- hibitors on yeast morphology were different. The number of morphological parameters commonly altered by these drugs was only approximately 6% of the total. Using a rational strategy to find commonly changed parameters,we focused on hidden essential similarities in the phenotypes possibly due to decreased ergosterol levels. This resulted in higher apparent morphological similarity. Improvements in morphological similarity were observed even when canonical correlation analysis was used to select biologically meaningful morphological parameters related to gene function. In addition to changes in cell morphology, we also observed differences in the synergistic effects among the three inhibitors and in their fungicidal effects against pathogenic fungi possibly due to the accumulation of different intermediates. This study provided a comprehensive understanding of the properties of inhibitors acting in the same biosynthetic pathway.

Assignment of unimodal probability distribution models for quantitative morphological phenotyping

Background

Cell morphology is a complex and integrative readout, and therefore, an attractive measurement for assessing the effects of genetic and chemical perturbations to cells. Microscopic images provide rich information on cell morphology; therefore, subjective morphological features are frequently extracted from digital images. However, measured datasets are fundamentally noisy; thus, estimation of the true values is an ultimate goal in quantitative morphological phenotyping. Ideal image analyses require precision, such as proper probability distribution analyses to detect subtle morphological changes, recall to minimize artifacts due to experimental error, and reproducibility to confirm the results.

Results

Here, we present UNIMO (UNImodal MOrphological data), a reliable pipeline for precise detection of subtle morphological changes by assigning unimodal probability distributions to morphological features of the budding yeast cells. By defining the data type, followed by validation using the model selection method, examination of 33 probability distributions revealed nine best-fitting probability distributions. The modality of the distribution was then clarified for each morphological feature using a probabilistic mixture model. Using a reliable and detailed set of experimental log data of wild-type morphological replicates, we considered the effects of confounding factors. As a result, most of the yeast morphological parameters exhibited unimodal distributions that can be used as basic tools for powerful downstream parametric analyses. The power of the proposed pipeline was confirmed by reanalyzing morphological changes in non-essential yeast mutants and detecting 1284 more mutants with morphological defects compared with a conventional approach (Box–Cox transformation). Furthermore, the combined use of canonical correlation analysis permitted global views on the cellular network as well as new insights into possible gene functions.

Conclusions

Based on statistical principles, we showed that UNIMO offers better predictions of the true values of morphological measurements. We also demonstrated how these concepts can provide biologically important information. This study draws attention to the necessity of employing a proper approach to do more with less.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12915-022-01283-6.

Genome-Wide DNA Methylation Analysis during Osteogenic Differentiation of Human Bone Marrow Mesenchymal Stem Cells

Bone marrow mesenchymal stem cells (BMSCs) nowadays are regarded as promising candidates in cell-based therapy for the regeneration of damaged bone tissues that are either incurable or intractable due to the insufficiency of current therapies. Recent studies suggest that BMSCs differentiate into osteoblasts, and that this differentiation is regulated by some specific patterns of epigenetic modifications, such as DNA methylation. However, the potential role of DNA methylation modification in BMSC osteogenic differentiation is unclear. In this study, we performed a genome-wide study of DNA methylation between the noninduced and induced osteogenic differentiation of BMSCs at day 7. We found that the majority of cytosines in a CpG context were methylated in induced BMSCs. Our results also revealed that, along with the induced osteogenic differentiation in BMSCs, the average genomic methylation levels and CpG methylation in transcriptional factor regions (TFs) were increased, the CpG methylation level of various genomic elements was mainly in the medium-high methylation section, and CpG methylation levels in the repeat element had highly methylated levels. The GO analysis of differentially methylated region- (DMR-) associated genes (DMGs) showed that GO terms, including cytoskeletal protein binding (included in Molecular Function GO terms), skeletal development (included in Biological Process GO terms), mesenchymal cell differentiation (included in Biological Process GO terms), and stem cell differentiation (included in Biological Process), were enriched in the hypermethylated DMGs. Then, the KEGG analysis results showed that the WNT pathway, inositol phosphate metabolism pathway, and cocaine addiction pathway were more correlative with the DMRs during the induced osteogenic differentiation in BMSCs. In conclusion, this study revealed the difference of methylated levels during the noninduced and induced osteogenic differentiation of BMSCs and provided useful information for future works to characterize the important function of epigenetic mechanisms on BMSCs' differentiation.

High-dimensional single-cell phenotyping reveals extensive haploinsufficiency

Haploinsufficiency, a dominant phenotype caused by a heterozygous loss-of-function mutation, has been rarely observed. However, high-dimensional single-cell phenotyping of yeast morphological characteristics revealed haploinsufficiency phenotypes for more than half of 1,112 essential genes under optimal growth conditions. Additionally, 40% of the essential genes with no obvious phenotype under optimal growth conditions displayed haploinsufficiency under severe growth conditions. Haploinsufficiency was detected more frequently in essential genes than in nonessential genes. Similar haploinsufficiency phenotypes were observed mostly in mutants with heterozygous deletion of functionally related genes, suggesting that haploinsufficiency phenotypes were caused by functional defects of the genes. A global view of the gene network was presented based on the similarities of the haploinsufficiency phenotypes. Our dataset contains rich information regarding essential gene functions, providing evidence that single-cell phenotyping is a powerful approach, even in the heterozygous condition, for analyzing complex biological systems.

Diploid organisms harboring a wild-type gene and a loss-of-function mutation are called heterozygotes. They are expected to have weak or no individual phenotypes because the mutation is compensated for by the intact allele. The dominant inheritance of phenotypes in heterozygotes is an exceptional phenomenon called haploinsufficiency. Haploinsufficiency was thought to be a rare occurrence; however, a sensitive technique called high-dimensional single-cell phenotyping challenges this perspective. Investigations of single-cell phenotypes revealed that a large extent of the essential genes in yeast exhibit haploinsufficiency. Our analyses also provided crucial information on gene functional networks based on haploinsufficiency phenotypes. This work shows that high-dimensional single-cell phenotyping is a useful tool that can be used to better understand complex biological systems.

Regulatory network analysis of hypertension and hypotension microarray data from mouse model

We aimed to identify the potential genes related to blood pressure regulation and screen target genes for high blood pressure (BPH) and low blood pressure (BPL) treatment. The GSE19817 microarray dataset, which included the aorta, liver, heart, and kidney samples from BPH, BPL, and normotensive mice, was downloaded from the Gene Expression Omnibus. Principal component analysis (PCA) was performed based on the entire expression profile. Differentially expressed genes (DEGs) were screened, followed by pathway enrichment analysis. Finally, gene regulatory networks were constructed based on BPH-related and BPL-related DEGs in the aorta, liver, heart, and kidney samples. As a result, DEGs were screened within their respective tissues due to high heterogeneity of different tissues. Totally, 2,726 BPH-related DEGs and 2,472 BPL-related DEGs were screened, which were mainly enriched in pathways such as immune response. The topology data of gene regulatory networks constructed by DEGs in the heart, kidney, and liver were similar than that in aorta. Finally, among BPH-related DEGs, Sept6 and Pigx were found in the top 10 differentially regulated DEGs by comparing the BPH-related DEGs of the aorta with the DEGs of the other 3 tissues in the regulatory network. Although among the top 10 differentially regulated BPL-related DEGs, no common differentially regulated DEGs were found, Wif1, Urb2, and Gtf2ird1 were found among the top ten DEGs in the three tissues other than the kidney tissue. Sept6 and Pigx might participate in the pathogenesis of BPH, whereas Gtf2ird1, Urb2, and Wif1 might be critical target genes for BPL treatment.

Characteristic gene selection via weighting principal components by singular values

Conventional gene selection methods based on principal component analysis (PCA) use only the first principal component (PC) of PCA or sparse PCA to select characteristic genes. These methods indeed assume that the first PC plays a dominant role in gene selection. However, in a number of cases this assumption is not satisfied, so the conventional PCA-based methods usually provide poor selection results. In order to improve the performance of the PCA-based gene selection method, we put forward the gene selection method via weighting PCs by singular values (WPCS). Because different PCs have different importance, the singular values are exploited as the weights to represent the influence on gene selection of different PCs. The ROC curves and AUC statistics on artificial data show that our method outperforms the state-of-the-art methods. Moreover, experimental results on real gene expression data sets show that our method can extract more characteristic genes in response to abiotic stresses than conventional gene selection methods.

Survival models with preclustered gene groups as covariates

Background

An important application of high dimensional gene expression measurements is the risk prediction and the interpretation of the variables in the resulting survival models. A major problem in this context is the typically large number of genes compared to the number of observations (individuals). Feature selection procedures can generate predictive models with high prediction accuracy and at the same time low model complexity. However, interpretability of the resulting models is still limited due to little knowledge on many of the remaining selected genes. Thus, we summarize genes as gene groups defined by the hierarchically structured Gene Ontology (GO) and include these gene groups as covariates in the hazard regression models. Since expression profiles within GO groups are often heterogeneous, we present a new method to obtain subgroups with coherent patterns. We apply preclustering to genes within GO groups according to the correlation of their gene expression measurements.

Results

We compare Cox models for modeling disease free survival times of breast cancer patients. Besides classical clinical covariates we consider genes, GO groups and preclustered GO groups as additional genomic covariates. Survival models with preclustered gene groups as covariates have similar prediction accuracy as models built only with single genes or GO groups.

Conclusions

The preclustering information enables a more detailed analysis of the biological meaning of covariates selected in the final models. Compared to models built only with single genes there is additional functional information contained in the GO annotation, and compared to models using GO groups as covariates the preclustering yields coherent representative gene expression profiles.

Discriminating classes of developmental toxicants using gene expression profiling in the embryonic stem cell test

The embryonic stem cell test (EST) has been shown to be a promising in vitro method for the prediction of developmental toxicity. In our previous studies, we demonstrated that the implementation of gene expression analysis in the EST may further improve the identification of developmental toxicants. In the present study, we investigated if gene expression profiling could be used to discriminate compound classes with distinct modes of action (MoA) using the EST protocol. Gene expression data of our previous study were used and were analyzed of embryonic stem cell (ESC) differentiation cultures exposed to six compounds belonging to two classes with distinct MoA, namely phthalates and triazoles. We used three approaches to study class-characteristic gene regulation that may be useful for discrimination of compound classes. First, at the individual gene level, gene signatures characteristic for each class were identified that successfully discriminated both classes using principal component analysis. Second, at the functional level, enriched gene ontology (GO) biological processes showed their usefulness for class discrimination via hierarchical clustering. Third, two previously identified gene sets, which we designed to predict developmental toxicity, appeared successful in separating phthalate from triazole compounds. In summary, we established the possibility to discriminate between compound classes in the EST system using three different specific transcriptomics-based approaches. Differential gene expression information specific for the class of compound under study may be employed to optimize prioritization of compounds within that class for further testing.

Concentration-dependent gene expression responses to flusilazole in embryonic stem cell differentiation cultures

The murine embryonic stem cell test (EST) is designed to evaluate developmental toxicity based on compound-induced inhibition of embryonic stem cell (ESC) differentiation into cardiomyocytes. The addition of transcriptomic evaluation within the EST may result in enhanced predictability and improved characterization of the applicability domain, therefore improving usage of the EST for regulatory testing strategies. Transcriptomic analyses assessing factors critical for risk assessment (i.e. dose) are needed to determine the value of transcriptomic evaluation in the EST. Here, using the developmentally toxic compound, flusilazole, we investigated the effect of compound concentration on gene expression regulation and toxicity prediction in ESC differentiation cultures. Cultures were exposed for 24 h to multiple concentrations of flusilazole (0.54-54 μM) and RNA was isolated. In addition, we sampled control cultures 0, 24, and 48 h to evaluate the transcriptomic status of the cultures across differentiation. Transcriptomic profiling identified a higher sensitivity of development-related processes as compared to cell division-related processes in flusilazole-exposed differentiation cultures. Furthermore, the sterol synthesis-related mode of action of flusilazole toxicity was detected. Principal component analysis using gene sets related to normal ESC differentiation was used to describe the dynamics of ESC differentiation, defined as the 'differentiation track'. The concentration-dependent effects on development were reflected in the significance of deviation of flusilazole-exposed cultures from this transcriptomic-based differentiation track. Thus, the detection of developmental toxicity in EST using transcriptomics was shown to be compound concentration-dependent. This study provides further insight into the possible application of transcriptomics in the EST as an improved alternative model system for developmental toxicity testing.