Quantitative proteomic analysis of rat liver for carcinogenicity prediction in a 28-day repeated dose study

The potential of quantitative proteomic analysis to predict carcinogenicity of chemical compounds was investigated. Using 2D-DIGE, we analyzed the effects of 63 chemical compounds on protein expression in the rat liver after 28 daily doses. Types of carcinogens were categorized depending on the species and organ specificity. The carcinogen characteristic proteins for each classification were identified by Welch's t value. For evaluation of the predictive concordance we used support vector machines. The rat hepatic carcinogen-specific classification gave higher concordance than the other classification. The generalization performance was measured by leave-one-out cross-validation. For genotoxic and non-genotoxic compounds, a concordance of 79.3 and 76.5%, respectively, was obtained by the top 30 ranked proteins with Welch's t value. Furthermore, we found that the increase of the expression level of the stress response proteins as the common feature of poorly predicted chemical compounds in the leave-20%-out cross-validation. Quantitative proteomics could be promising technique for developing biomarker panels that can be used for carcinogenicity prediction. The list of proteins identified in this study and the zoomed gel images of the top ranked proteins in statistic analysis are provided in Supplementary Data.     

A bayesian mixed regression based prediction of quantitative traits from molecular marker and gene expression data

Both molecular marker and gene expression data were considered alone as well as jointly to serve as additive predictors for two pathogen-activity-phenotypes in real recombinant inbred lines of soybean. For unobserved phenotype prediction, we used a bayesian hierarchical regression modeling, where the number of possible predictors in the model was controlled by different selection strategies tested. Our initial findings were submitted for DREAM5 (the 5th Dialogue on Reverse Engineering Assessment and Methods challenge) and were judged to be the best in sub-challenge B3 wherein both functional genomic and genetic data were used to predict the phenotypes. In this work we further improve upon this previous work by considering various predictor selection strategies and cross-validation was used to measure accuracy of in-data and out-data predictions. The results from various model choices indicate that for this data use of both data types (namely functional genomic and genetic) simultaneously improves out-data prediction accuracy. Adequate goodness-of-fit can be easily achieved with more complex models for both phenotypes, since the number of potential predictors is large and the sample size is not small. We also further studied gene-set enrichment (for continuous phenotype) in the biological process in question and chromosomal enrichment of the gene set. The methodological contribution of this paper is in exploration of variable selection techniques to alleviate the problem of over-fitting. Different strategies based on the nature of covariates were explored and all methods were implemented under the bayesian hierarchical modeling framework with indicator-based covariate selection. All the models based in careful variable selection procedure were found to produce significant results based on permutation test.     

Accounting for Genetic Architecture Improves Sequence Based Genomic Prediction for a Drosophila Fitness Trait

The ability to predict quantitative trait phenotypes from molecular polymorphism data will revolutionize evolutionary biology, medicine and human biology, and animal and plant breeding. Efforts to map quantitative trait loci have yielded novel insights into the biology of quantitative traits, but the combination of individually significant quantitative trait loci typically has low predictive ability. Utilizing all segregating variants can give good predictive ability in plant and animal breeding populations, but gives little insight into trait biology. Here, we used the Drosophila Genetic Reference Panel to perform both a genome wide association analysis and genomic prediction for the fitness-related trait chill coma recovery time. We found substantial total genetic variation for chill coma recovery time, with a genetic architecture that differs between males and females, a small number of molecular variants with large main effects, and evidence for epistasis. Although the top additive variants explained 36% (17%) of the genetic variance among lines in females (males), the predictive ability using genomic best linear unbiased prediction and a relationship matrix using all common segregating variants was very low for females and zero for males. We hypothesized that the low predictive ability was due to the mismatch between the infinitesimal genetic architecture assumed by the genomic best linear unbiased prediction model and the true genetic architecture of chill coma recovery time. Indeed, we found that the predictive ability of the genomic best linear unbiased prediction model is markedly improved when we combine quantitative trait locus mapping with genomic prediction by only including the top variants associated with main and epistatic effects in the relationship matrix. This trait-associated prediction approach has the advantage that it yields biologically interpretable prediction models.     

Potent immunogenicity in BRCA1‐mutated patients with high‐grade serous ovarian carcinoma

High‐grade serous ovarian carcinomas (HGSOCs) were among the tumours with an unsatisfactory outcome of immune checkpoint inhibitors (ICIs). It is imperative to develop feasible biomarker for identifying responsive candidates and guiding precise immunotherapy for HGSOC patients. Here, we analysed genomic data of patients with HGSOCs to depict their immunological phenotype of tumour microenvironment (TME) and figure out the major determinants of immunogenicity. In comparison with other solid tumours, we observed the lowest levels of PD‐L1, total mutation burden (TMB) and cytolytic molecules in HGSOCs. Surprisingly, TMB is not certainly positively related to tumour immune response as it failed to predict the response to ICIs in a considerable portion of patients in previous clinical trials. By a machine learning approach in search of biomarkers for immunotherapy implications for HGSOCs, we identified the ten most dominant factors determining the immunogenicity of HGSOCs. Interestingly, we found that BRCA1 mutated tumours presented a potent immunogenic phenotype, independent of TMB, meeting the criteria of both our dominant factors and the determinants of immunogenicity established before. Our findings provide evidence that BRCA1‐mutation may be served as a predictive biomarker in guiding ICI therapies for the patients with HGSOCs.     

How to Infer Relative Fitness from a Sample of Genomic Sequences

Mounting evidence suggests that natural populations can harbor extensive fitness diversity with numerous genomic loci under selection. It is also known that genealogical trees for populations under selection are quantifiably different from those expected under neutral evolution and described statistically by Kingman’s coalescent. While differences in the statistical structure of genealogies have long been used as a test for the presence of selection, the full extent of the information that they contain has not been exploited. Here we demonstrate that the shape of the reconstructed genealogical tree for a moderately large number of random genomic samples taken from a fitness diverse, but otherwise unstructured, asexual population can be used to predict the relative fitness of individuals within the sample. To achieve this we define a heuristic algorithm, which we test in silico, using simulations of a Wright–Fisher model for a realistic range of mutation rates and selection strength. Our inferred fitness ranking is based on a linear discriminator that identifies rapidly coalescing lineages in the reconstructed tree. Inferred fitness ranking correlates strongly with actual fitness, with a genome in the top 10% ranked being in the top 20% fittest with false discovery rate of 0.1–0.3, depending on the mutation/selection parameters. The ranking also enables us to predict the genotypes that future populations inherit from the present one. While the inference accuracy increases monotonically with sample size, samples of 200 nearly saturate the performance. We propose that our approach can be used for inferring relative fitness of genomes obtained in single-cell sequencing of tumors and in monitoring viral outbreaks.