Mirnacle: machine learning with SMOTE and random forest for improving selectivity in pre-miRNA ab initio prediction.

BACKGROUND: MicroRNAs (miRNAs) are key gene expression regulators in plants and animals. Therefore, miRNAs are involved in several biological processes, making the study of these molecules one of the most relevant topics of molecular biology nowadays. However, characterizing miRNAs in vivo is still a complex task. As a consequence, in silico methods have been developed to predict miRNA loci. A common ab initio strategy to find miRNAs in genomic data is to search for sequences that can fold into the typical hairpin structure of miRNA precursors (pre-miRNAs). The current ab initio approaches, however, have selectivity issues, i.e., a high number of false positives is reported, which can lead to laborious and costly attempts to provide biological validation. This study presents an extension of the ab initio method miRNAFold, with the aim of improving selectivity through machine learning techniques, namely, random forest combined with the SMOTE procedure that copes with imbalance datasets. RESULTS: By comparing our method, termed Mirnacle, with other important approaches in the literature, we demonstrate that Mirnacle substantially improves selectivity without compromising sensitivity. For the three datasets used in our experiments, our method achieved at least 97% of sensitivity and could deliver a two-fold, 20-fold, and 6-fold increase in selectivity, respectively, compared with the best results of current computational tools. CONCLUSIONS: The extension of miRNAFold by the introduction of machine learning techniques, significantly increases selectivity in pre-miRNA ab initio prediction, which optimally contributes to advanced studies on miRNAs, as the need of biological validations is diminished. Hopefully, new research, such as studies of severe diseases caused by miRNA malfunction, will benefit from the proposed computational tool.

The draft genomes of Crassostrea gasar and Crassostrea rhizophorae: key resources for leveraging oyster cultivation in the Southwest Atlantic.

OBJECTIVES: The two oyster species studied hold considerable economic importance for artisanal harvest (Crassostrea rhizophorae) and aquaculture (Crassostrea gasar). Their draft genomes will play an important role in the application of genomic methods such as RNAseq, population-based genomic scans aiming at addressing expression responses to pollution stress, adaptation to salinity and temperature variation, and will also permit investigating the genetic bases and enable marker-assisted selection of economically important traits like shell and mantle coloration and resistance to temperature and disease. DATA DESCRIPTION: The draft assembly size of Crassostrea gasar is 506 Mbp, and of Crassostrea rhizophorae is 584 Mbp with scaffolds N50 of 11,3 Mbp and 4,9 Mbp, respectively. The general masked bases by RepeatMasker in both genomes were highly similar using different datasets. The masked bases varied from 9.41% in C. gasar to 10.05% in C. rhizophorae and 42.85% in C. gasar to 44.44% in C. rhizophorae using Dfam and RepeatModeler datasets, respectively. Functional annotation with eggNog resulted in 34,693 annotated proteins in C. rhizophorae and 26,328 in C. gasar. BUSCO analysis shows that almost 99% of genes (5,295) are complete in relation to the mollusk orthologous genes dataset (mollusca_odb10).

Genomic basis of broad host range and environmental adaptability of Rhizobium tropici CIAT 899 and Rhizobium sp. PRF 81 which are used in inoculants for common bean (Phaseolus vulgaris L.).

BACKGROUND: Rhizobium tropici CIAT 899 and Rhizobium sp. PRF 81 are α-Proteobacteria that establish nitrogen-fixing symbioses with a range of legume hosts. These strains are broadly used in commercial inoculants for application to common bean (Phaseolus vulgaris) in South America and Africa. Both strains display intrinsic resistance to several abiotic stressful conditions such as low soil pH and high temperatures, which are common in tropical environments, and to several antimicrobials, including pesticides. The genetic determinants of these interesting characteristics remain largely unknown. RESULTS: Genome sequencing revealed that CIAT 899 and PRF 81 share a highly-conserved symbiotic plasmid (pSym) that is present also in Rhizobium leucaenae CFN 299, a rhizobium displaying a similar host range. This pSym seems to have arisen by a co-integration event between two replicons. Remarkably, three distinct nodA genes were found in the pSym, a characteristic that may contribute to the broad host range of these rhizobia. Genes for biosynthesis and modulation of plant-hormone levels were also identified in the pSym. Analysis of genes involved in stress response showed that CIAT 899 and PRF 81 are well equipped to cope with low pH, high temperatures and also with oxidative and osmotic stresses. Interestingly, the genomes of CIAT 899 and PRF 81 had large numbers of genes encoding drug-efflux systems, which may explain their high resistance to antimicrobials. Genome analysis also revealed a wide array of traits that may allow these strains to be successful rhizosphere colonizers, including surface polysaccharides, uptake transporters and catabolic enzymes for nutrients, diverse iron-acquisition systems, cell wall-degrading enzymes, type I and IV pili, and novel T1SS and T5SS secreted adhesins. CONCLUSIONS: Availability of the complete genome sequences of CIAT 899 and PRF 81 may be exploited in further efforts to understand the interaction of tropical rhizobia with common bean and other legume hosts.