Enhancing the Resolution of Rumen Microbial Classification from Metatranscriptomic Data Using Kraken and Mothur

Abstract

The advent of next generation sequencing and bioinformatics tools have greatly advanced our knowledge about the phylogenetic diversity and ecological role of microbes inhabiting the mammalian gut. However, there is a lack of information on the evaluation of these computational tools in the context of the rumen microbiome as these programs have mostly been benchmarked on real or simulated datasets generated from human studies. In this study, we compared the outcomes of two methods, Kraken (mRNA based) and a pipeline developed in-house based on Mothur (16S rRNA based), to assess the taxonomic profiles (bacteria and archaea) of rumen microbial communities using total RNA sequencing of rumen fluid collected from 12 cattle with differing feed conversion ratios (FCR). Both approaches revealed a similar phyla distribution of the most abundant taxa, with Bacteroidetes, Firmicutes, and Proteobacteria accounting for approximately 80% of total bacterial abundance. For bacterial taxa, although 69 genera were commonly detected by both methods, an additional 159 genera were exclusively identified by Kraken. Kraken detected 423 species, while Mothur was not able to assign bacterial sequences to the species level. For archaea, both methods generated similar results only for the abundance of Methanomassiliicoccaceae (previously referred as RCC), which comprised more than 65% of the total archaeal families. Taxon R4-41B was exclusively identified by Mothur in the rumen of feed efficient bulls, whereas Kraken uniquely identified Methanococcaceae in inefficient bulls. Although Kraken enhanced the microbial classification at the species level, identification of bacteria or archaea in the rumen is limited due to a lack of reference genomes for the rumen microbiome. The findings from this study suggest that the development of the combined pipelines using Mothur and Kraken is needed for a more inclusive and representative classification of microbiomes.

A Comparison of Two DNA Metagenomic Bioinformatic Pipelines While Evaluating the Microbial Diversity in Feces of Tanzanian Small Holder Dairy Cattle

“…Recent investigations have shown gut Cyanobacteria to be very conserved, but their 16S rRNA phylogenetic tree was different from the photosynthetic ones; this has led to the designation of a new putative class called Melainabacteria [47], whose members were able to ferment a variety of sugars in the gut [48]. Similar to other previous studies [21,39] neither Kraken 2 nor MG-RAST identified Melainabacteria in the samples, showing the need for further research to distinguish their role in the gut of cattle.…”

“…Cyanobacteria can be both heterocystous and nonheterocystous [43]. Although the ruminal environment is commonly deemed anaerobic, significant levels of oxygen in the rumen fluid can be identified [44], suggesting that the occurrence of cyanobacteria in the rumen may be associated with the scavenging of oxygen and the fermentation of sugar under restricted aerobic environments [21]. Whereas Cyanobacteria has been extensively detected in aqueous and soil environments [45,46], it is important to point out that the identification of this phylum in the gut of humans raises grave questions regarding their role in aphotic and anaerobic habitats like the rumen [47].…”

“…Identification of this bacterial species may show that Kraken 2 did not correctly identify the microbe, since the reference genome data was based mostly on all microbial genomes annotated in the NCBI database. These organisms may have been identified in the rumen, however, since cattle can eat soil, which makes it possible to detect them temporarily [21].…”

“…This may be attributed to the complexity of the microbial communities that consist of archaea, bacteria, fungi, and protozoa [20]. For example, a previous research by [21] compared taxonomic compositions of rumen microbial communities using Kraken 2 and an in-house pipeline developed based on MOTHUR to compare the rumen fluid RNA collected from cattle with different feed conversion ratios (FCR). The study found out that a similar distribution of the most abundant taxa was found in both pipelines at the phylum level; however, unlike raken 2, MOTHUR was unable to assign sequences to the species level while Kraken 2's ability to identify microbes was restricted due to an absence of some rumen microbiome reference genomes.…”

Rumen Bacteria and Serum Metabolites Predictive of Feed Efficiency Phenotypes in Beef Cattle

“…Cyanobacteria was identified as a class with low-ranked and non-fluctuating proportions across all samples and used as the denominator in the log-ratio. Although, Cyanobacteria are oxygenic phototrophic bacteria they are often found at low but constant proportions in the rumen 24,25 and thought to be possibly misclassified from the class Melainabacteria 26 . By using the log-ratio of Flavobacteriia and Cyanobacteria, the assumption can be made that this correlation is caused by an enrichment in Flavobacteriia and not a decrease in Cyanobacteria.…”