A Case of Gene Fragmentation in Plant Mitochondria Fixed by the Selection of a Compensatory Restorer of Fertility-Like PPR Gene.

The high mutational load of mitochondrial genomes combined with their uniparental inheritance and high polyploidy favors the maintenance of deleterious mutations within populations. How cells compose and adapt to the accumulation of disadvantageous mitochondrial alleles remains unclear. Most harmful changes are likely corrected by purifying selection, however, the intimate collaboration between mitochondria- and nuclear-encoded gene products offers theoretical potential for compensatory adaptive changes. In plants, cytoplasmic male sterilities are known examples of nucleo-mitochondrial coadaptation situations in which nuclear-encoded restorer of fertility (Rf) genes evolve to counteract the effect of mitochondria-encoded cytoplasmic male sterility (CMS) genes and restore fertility. Most cloned Rfs belong to a small monophyletic group, comprising 26 pentatricopeptide repeat genes in Arabidopsis, called Rf-like (RFL). In this analysis, we explored the functional diversity of RFL genes in Arabidopsis and found that the RFL8 gene is not related to CMS suppression but essential for plant embryo development. In vitro-rescued rfl8 plantlets are deficient in the production of the mitochondrial heme-lyase complex. A complete ensemble of molecular and genetic analyses allowed us to demonstrate that the RFL8 gene has been selected to permit the translation of the mitochondrial ccmFN2 gene encoding a heme-lyase complex subunit which derives from the split of the ccmFN gene, specifically in Brassicaceae plants. This study represents thus a clear case of nuclear compensation to a lineage-specific mitochondrial genomic rearrangement in plants and demonstrates that RFL genes can be selected in response to other mitochondrial deviancies than CMS suppression.

Identification of a Transcriptomic Prognostic Signature by Machine Learning Using a Combination of Small Cohorts of Prostate Cancer.

Determining which treatment to provide to men with prostate cancer (PCa) is a major challenge for clinicians. Currently, the clinical risk-stratification for PCa is based on clinico-pathological variables such as Gleason grade, stage and prostate specific antigen (PSA) levels. But transcriptomic data have the potential to enable the development of more precise approaches to predict evolution of the disease. However, high quality RNA sequencing (RNA-seq) datasets along with clinical data with long follow-up allowing discovery of biochemical recurrence (BCR) biomarkers are small and rare. In this study, we propose a machine learning approach that is robust to batch effect and enables the discovery of highly predictive signatures despite using small datasets. Gene expression data were extracted from three RNA-Seq datasets cumulating a total of 171 PCa patients. Data were re-analyzed using a unique pipeline to ensure uniformity. Using a machine learning approach, a total of 14 classifiers were tested with various parameters to identify the best model and gene signature to predict BCR. Using a random forest model, we have identified a signature composed of only three genes (JUN, HES4, PPDPF) predicting BCR with better accuracy [74.2%, balanced error rate (BER) = 27%] than the clinico-pathological variables (69.2%, BER = 32%) currently in use to predict PCa evolution. This score is in the range of the studies that predicted BCR in single-cohort with a higher number of patients. We showed that it is possible to merge and analyze different small and heterogeneous datasets altogether to obtain a better signature than if they were analyzed individually, thus reducing the need for very large cohorts. This study demonstrates the feasibility to regroup different small datasets in one larger to identify a predictive genomic signature that would benefit PCa patients.

Immune-focused multi-omics analysis of prostate cancer: leukocyte Ig-Like receptors are associated with disease progression.

Prostate cancer (PCa) immunotherapy has shown limited efficacy so far, even in advanced-stage cancers. The success rate of PCa immunotherapy might be improved by approaches more adapted to the immunobiology of the disease. The objective of this study was to perform a multi-omics analysis to identify immune genes associated with PCa progression to better characterize PCa immunobiology and propose new immunotherapeutic targets. mRNA, miRNA, methylation, copy number aberration, and single nucleotide variant datasets from The Cancer Genome Atlas PRAD cohort were analyzed after filtering for genes associated with immunity. Sparse partial least squares-discriminant analyses were performed to identify features associated with biochemical recurrence (BCR) in each type of omics data. Selected features predicted BCR with a balanced error rate (BER) of 0.20 to 0.51 in single-omics and of 0.05 in multi-omics analyses. Amongst features associated with BCR were genes from the Immunoglobulin Ig-like Receptor (LILR) family which are immune checkpoints with immunotherapeutic potential. Using Multivariate INTegrative (MINT) analysis, the association of five LILR genes with BCR was quantified in a combination of three RNA-seq datasets and confirmed with Kaplan-Meier analysis in both these and in an independent RNA-seq dataset. Finally, immunohistochemistry showed that a high number of LILRB1 positive cells within the tumors predicted long-term adverse outcomes. Thus, tumors characterized by abnormal expression of LILR genes have an elevated risk of recurring after definitive local therapy. The immunotherapeutic potential of these regulators to stimulate the immune response against PCa should be evaluated in pre-clinical models.

Sampling uncertainty versus method uncertainty: A general framework with applications to omics biomarker selection.

Uncertainty is a crucial issue in statistics which can be considered from different points of view. One type of uncertainty, typically referred to as sampling uncertainty, arises through the variability of results obtained when the same analysis strategy is applied to different samples. Another type of uncertainty arises through the variability of results obtained when using the same sample but different analysis strategies addressing the same research question. We denote this latter type of uncertainty as method uncertainty. It results from all the choices to be made for an analysis, for example, decisions related to data preparation, method choice, or model selection. In medical sciences, a large part of omics research is focused on the identification of molecular biomarkers, which can either be performed through ranking or by selection from among a large number of candidates. In this paper, we introduce a general resampling-based framework to quantify and compare sampling and method uncertainty. For illustration, we apply this framework to different scenarios related to the selection and ranking of omics biomarkers in the context of acute myeloid leukemia: variable selection in multivariable regression using different types of omics markers, the ranking of biomarkers according to their predictive performance, and the identification of differentially expressed genes from RNA-seq data. For all three scenarios, our findings suggest highly unstable results when the same analysis strategy is applied to two independent samples, indicating high sampling uncertainty and a comparatively smaller, but non-negligible method uncertainty, which strongly depends on the methods being compared.

Transcriptional and metabolic adjustments in ADP-glucose pyrophosphorylase-deficient bt2 maize kernels.

During the cloning of monogenic recessive mutations responsible for a defective kernel phenotype in a Mutator-induced Zea mays mutant collection, we isolated a new mutant allele in Brittle2 (Bt2), which codes for the small subunit of ADP-glucose pyrophosphorylase (AGPase), a key enzyme in starch synthesis. Reverse transcription-polymerase chain reaction experiments with gene-specific primers confirmed a predominant expression of Bt2 in endosperm, of Agpsemzm in embryo, and of Agpslzm in leaf, but also revealed considerable additional expression in various tissues for all three genes. Bt2a, the classical transcript coding for a cytoplasmic isoform, was almost exclusively expressed in the developing endosperm, whereas Bt2b, an alternative transcript coding for a plastidial isoform, was expressed in almost all tissues tested with a pattern very similar to that of Agpslzm. The phenotypic analysis showed that, at 30 d after pollination (DAP), mutant kernels were plumper than wild-type kernels, that the onset of kernel collapse took place between 31 and 35 DAP, and that the number of starch grains was greatly reduced in the mutant endosperm but not the mutant embryo. A comparative transcriptome analysis of wild-type and bt2-H2328 kernels at middevelopment (35 DAP) with the 18K GeneChip Maize Genome Array led to the conclusion that the lack of Bt2-encoded AGPase triggers large-scale changes on the transcriptional level that concern mainly genes involved in carbohydrate or amino acid metabolic pathways. Principal component analysis of (1)H nuclear magnetic resonance metabolic profiles confirmed the impact of the bt2-H2328 mutation on these pathways and revealed that the bt2-H2328 mutation did not only affect the endosperm, but also the embryo at the metabolic level. These data suggest that, in the bt2-H2328 endosperms, regulatory networks are activated that redirect excess carbon into alternative biosynthetic pathways (amino acid synthesis) or into other tissues (embryo).

Genome-wide expression profiling of the host response to root-knot nematode infection in Arabidopsis.

During a compatible interaction, root-knot nematodes (Meloidogyne spp.) induce the redifferentiation of root cells into multinucleate nematode feeding cells (giant cells). Hyperplasia and hypertrophy of the surrounding cells leads to the formation of a root gall. We investigated the plant response to root-knot nematodes by carrying out a global analysis of gene expression during gall formation in Arabidopsis, using giant cell-enriched root tissues. Among 22 089 genes monitored with the complete Arabidopsis transcriptome microarray gene-specific tag, we identified 3373 genes that display significant differential expression between uninfected root tissues and galls at different developmental stages. Quantitative PCR analysis and the use of promoter GUS fusions confirmed the changes in mRNA levels observed in our microarray analysis. We showed that a comparable number of genes were found to be up- and downregulated, indicating that gene downregulation might be essential to allow proper gall formation. Moreover, many genes belonging to the same family are differently regulated in feeding cells. This genome-wide overview of gene expression during plant-nematode interaction provides new insights into nematode feeding-cell formation, and highlights that the suppression of plant defence is associated with nematode feeding-site development.