A practical introduction to Random Forest for genetic association studies in ecology and evolution

Large genomic studies are becoming increasingly common with advances in sequencing technology, and our ability to understand how genomic variation influences phenotypic variation between individuals has never been greater. The exploration of such relationships first requires the identification of associations between molecular markers and phenotypes. Here, we explore the use of Random Forest (RF), a powerful machine‐learning algorithm, in genomic studies to discern loci underlying both discrete and quantitative traits, particularly when studying wild or nonmodel organisms. RF is becoming increasingly used in ecological and population genetics because, unlike traditional methods, it can efficiently analyse thousands of loci simultaneously and account for nonadditive interactions. However, understanding both the power and limitations of Random Forest is important for its proper implementation and the interpretation of results. We therefore provide a practical introduction to the algorithm and its use for identifying associations between molecular markers and phenotypes, discussing such topics as data limitations, algorithm initiation and optimization, as well as interpretation. We also provide short R tutorials as examples, with the aim of providing a guide to the implementation of the algorithm. Topics discussed here are intended to serve as an entry point for molecular ecologists interested in employing Random Forest to identify trait associations in genomic data sets.     

An integrated linkage map reveals candidate genes underlying adaptive variation in Chinook salmon (Oncorhynchus tshawytscha)

Salmonids are an important cultural and ecological resource exhibiting near worldwide distribution between their native and introduced range. Previous research has generated linkage maps and genomic resources for several species as well as genome assemblies for two species. We first leveraged improvements in mapping and genotyping methods to create a dense linkage map for Chinook salmon Oncorhynchus tshawytscha by assembling family data from different sources. We successfully mapped 14 620 SNP loci including 2336 paralogs in subtelomeric regions. This improved map was then used as a foundation to integrate genomic resources for gene annotation and population genomic analyses. We anchored a total of 286 scaffolds from the Atlantic salmon genome to the linkage map to provide a framework for the placement 11 728 Chinook salmon ESTs. Previously identified thermotolerance QTL were found to colocalize with several candidate genes including HSP70, a gene known to be involved in thermal response, as well as its inhibitor. Multiple regions of the genome with elevated divergence between populations were also identified, and annotation of ESTs in these regions identified candidate genes for fitness related traits such as stress response, growth and behaviour. Collectively, these results demonstrate the utility of combining genomic resources with linkage maps to enhance evolutionary inferences.     

Long‐term effect of biochar on the stabilization of recent carbon: soils with historical inputs of charcoal

This study was set up to identify the long‐term effect of biochar on soil C sequestration of recent carbon inputs. Arable fields (n = 5) were found in Belgium with charcoal‐enriched black spots (>50 m²; n = 14) dating >150 years ago from historical charcoal production mound kilns. Topsoils from these ‘black spots’ had a higher organic C concentration [3.6 ± 0.9% organic carbon (OC)] than adjacent soils outside these black spots (2.1 ± 0.2% OC). The soils had been cropped with maize for at least 12 years which provided a continuous input of C with a C isotope signature (δ¹³C) ?13.1, distinct from the δ¹³C of soil organic carbon (?27.4 ‰) and charcoal (?25.7 ‰) collected in the surrounding area. The isotope signatures in the soil revealed that maize‐derived C concentration was significantly higher in charcoal‐amended samples (‘black spots’) than in adjacent unamended ones (0.44% vs. 0.31%; P = 0.02). Topsoils were subsequently collected as a gradient across two ‘black spots’ along with corresponding adjacent soils outside these black spots and soil respiration, and physical soil fractionation was conducted. Total soil respiration (130 days) was unaffected by charcoal, but the maize‐derived C respiration per unit maize‐derived OC in soil significantly decreased about half (P < 0.02) with increasing charcoal‐derived C in soil. Maize‐derived C was proportionally present more in protected soil aggregates in the presence of charcoal. The lower specific mineralization and increased C sequestration of recent C with charcoal are attributed to a combination of physical protection, C saturation of microbial communities and, potentially, slightly higher annual primary production. Overall, this study provides evidence of the capacity of biochar to enhance C sequestration in soils through reduced C turnover on the long term.     

16S ribosomal DNA-directed PCR primers for ruminal methanogens and identification of methanogens colonising young lambs

The population densities and identities of methanogens colonising new-born lambs in a grazing flock were determined from rumen samples collected at regular intervals after birth. Methanogen colonisation was found at the first sampling (1-3 days after birth) and population densities reached around 10(4) methanogens per gram at 1 week of age. Population densities increased in an exponential manner to a maximum of 10(8)-10(9) per gram at 3 weeks of age. To identify methanogens, PCR primers specific for each of the Archaea; a grouping of the orders Methanomicrobiales, Methanosarcinales and Methanococcales; the order Methanobacteriales; the order Methanococcales; the order Methanosarcinales; the genus Methanobacterium; and the genus Methanobrevibacter were designed. Primer-pair specificities were confirmed in tests with target and non-target micro-organisms. PCR analysis of DNA extracts revealed that all the detectable ruminal methanogens belonged to the order Methanobacteriales, with no methanogens belonging to the Methanomicrobiales, the Methanosarcinales, or the Methanococcales being detected. In 3 lambs, the initial colonising methanogens were Methanobrevibacter spp. and in 2 lambs were a mixture of Methanobrevibacter and Methanobacterium spp. In the latter case, the initial colonising Methanobacterium spp. subsequently disappeared and were not detectable 12-19 days after birth. Seven weeks after birth, lambs contained only Methanobrevibacter spp. This study, the first to provide information on the identities of methanogens colonising pre-ruminants, suggests that the predominant methanogens found in the mature rumen establish very soon after birth and well before a functioning rumen develops.     

Integration of Random Forest with population‐based outlier analyses provides insight on the genomic basis and evolution of run timing in Chinook salmon (Oncorhynchus tshawytscha)

Anadromous Chinook salmon populations vary in the period of river entry at the initiation of adult freshwater migration, facilitating optimal arrival at natal spawning. Run timing is a polygenic trait that shows evidence of rapid parallel evolution in some lineages, signifying a key role for this phenotype in the ecological divergence between populations. Studying the genetic basis of local adaptation in quantitative traits is often impractical in wild populations. Therefore, we used a novel approach, Random Forest, to detect markers linked to run timing across 14 populations from contrasting environments in the Columbia River and Puget Sound, USA. The approach permits detection of loci of small effect on the phenotype. Divergence between populations at these loci was then examined using both principle component analysis and F_ST outlier analyses, to determine whether shared genetic changes resulted in similar phenotypes across different lineages. Sequencing of 9107 RAD markers in 414 individuals identified 33 predictor loci explaining 79.2% of trait variance. Discriminant analysis of principal components of the predictors revealed both shared and unique evolutionary pathways in the trait across different lineages, characterized by minor allele frequency changes. However, genome mapping of predictor loci also identified positional overlap with two genomic outlier regions, consistent with selection on loci of large effect. Therefore, the results suggest selective sweeps on few loci and minor changes in loci that were detected by this study. Use of a polygenic framework has provided initial insight into how divergence in a trait has occurred in the wild.     

Quantitative Determination of H2-Utilizing Acetogenic and Sulfate-Reducing Bacteria and Methanogenic Archaea from Digestive Tract of Different Mammals

Total number of bacteria, cellulolytic bacteria, and H₂-utilizing microbial populations (methanogenic archaea, acetogenic and sulfate-reducing bacteria) were enumerated in fresh rumen samples from sheep, cattle, buffaloes, deer, llamas, and caecal samples from horses. Methanogens and sulfate reducers were found in all samples, whereas acetogens were not detected in some samples of each animal. Archaea methanogens were the largest H₂-utilizing populations in all animals, and a correlation was observed between the numbers of methanogens and those of cellulolytic microorganisms. Higher counts of acetogens were found in horses and llamas (1 × 10⁴ and 4 × 10⁴ cells ml⁻¹ respectively).     

Energy conversion in an internally illuminated annular‐plate airlift photobioreactor

Algal‐derived therapeutics, bioactive molecules, and fuels produced in photobioreactors (PBRs) are of great scientific and economic interest, but the high cost of production still prevents their widespread use. Specifically, the cost of the energy inputs and the control of the photonic inputs that enable production optimization continue to be problematic. To this end, a novel 55‐L annular‐plate airlift PBR (APAPBR) with internal illumination was designed and characterized for the batch production of algal biomass. The APAPBR was able to convert mixing and photonic energy inputs into Chlorella pituita SG1 biomass at an efficiency of 0.064 (J biomass [J input]^?1), or 0.27 g dry cell weight (DW) W^?1 d^?1. Thanks to a high degree of photon capture and the airlift effect that provided energy‐efficient mixing and mass transfer, this energy conversion is 54% of the theoretical maximum as determined in previous studies. Under these efficiency conditions, C. pituita SG1 was able to grow photoautotrophically to 3.9 ± 0.2 gDW L^?1. Additionally, a mathematical approach was used to predict the mean light intensity with the highest biomass yield per unit of photonic input and the maximum biomass concentration achievable under the given process conditions. These predictions were validated in our system by the experimental cultivation data. This APAPBR represents a simple, innovative, and energy‐efficient PBR configuration that could decrease the cost of phototrophic bioprocesses and enable novel bioprocesses that require a high degree of control over the photonic input.     

Hydration of Oleic Acid by Enterococcus gallinarum, Pediococcus acidilactici andLactobacillus sp. Isolated from the Rumen

Three ruminal bacteria found to convert oleic acid to 10-hydroxystearic acid were identified asEnterococcus gallinarum , Pediococcus acidilactici and a lactobacillus physiologically similar toLactobacillus reuteri and L. fermentum. The oleic acid-hydrating properties of the three bacteria were determined and hydration was shown to be predominantly a feature of anaerobic culture with relatively little hydration occurring in aerobic culture. Hydration was highly pH dependent and not related to cell growth. At the optimal hydration pH for each bacterium, hydration yields were 97%, 93% and 76% for the E. gallinarum, P. acidilactici andLactobacillus strains, respectively. The enterococcus and the lactobacillus hydrated oleic acid only after growth ceased whereas the pediococcus hydrated oleic acid during the late logarithmic growth. Hydration was not specific for oleic acid with all three bacteria hydrating the oleic acid homologue palmitoleic acid. None of the bacteria hydrogenated oleic acid to stearic acid. Our results suggest that a capacity to hydrate oleic acid may be a property of many lactic acid bacteria.     

Hydration of linoleic acid by bacteria isolated from ruminants

Two strains of Enterococcus faecalis isolated from the ovine rumen and known to hydrate oleic acid were shown to transform linoleic acid by hydration into two products. The products, identified as 10-hydroxy-12-octadecenoic acid and 13-hydroxy-9-octadecenoic acid, were formed during stationary phase in yields of 13% and 6% respectively. Yields increased to 22% and 14% when culture conditions were optimised. To our knowledge, this is the first report of 13-hydroxy-9-octadecenoic acid production by bacteria. During a search for further linoleic-acid-hydrating bacteria, a strain of Streptococcus bovis isolated from bovine faeces and the ruminal strain S. bovis JB1 were found to hydrate linoleic acid. Both strains formed only one product and the most rapid appearance occurred during exponential growth. The S. bovis product, identified as 13-hydroxy-9-octadecenoic acid, formed in a yield of 28%. This study provides the first information on linoleic acid hydration by ruminal bacteria.