Predicting the HMA-LMA Status in Marine Sponges by Machine Learning.

Lucas Moitinho-Silva,Torsten Thomas,Yu-Chen Wu,Roman Marchant,Nicole Webster,Susanna López-Legentil,Ute Hentschel,Shaun Nielsen,Grace P Mccormack,Georg Steinert,Cristiane C P Hardoim

doi:10.3389/fmicb.2017.00752

Abstract

The dichotomy between high microbial abundance (HMA) and low microbial abundance (LMA) sponges has been observed in sponge-microbe symbiosis, although the extent of this pattern remains poorly unknown. We characterized the differences between the microbiomes of HMA (n = 19) and LMA (n = 17) sponges (575 specimens) present in the Sponge Microbiome Project. HMA sponges were associated with richer and more diverse microbiomes than LMA sponges, as indicated by the comparison of alpha diversity metrics. Microbial community structures differed between HMA and LMA sponges considering Operational Taxonomic Units (OTU) abundances and across microbial taxonomic levels, from phylum to species. The largest proportion of microbiome variation was explained by the host identity. Several phyla, classes, and OTUs were found differentially abundant in either group, which were considered “HMA indicators” and “LMA indicators.” Machine learning algorithms (classifiers) were trained to predict the HMA-LMA status of sponges. Among nine different classifiers, higher performances were achieved by Random Forest trained with phylum and class abundances. Random Forest with optimized parameters predicted the HMA-LMA status of additional 135 sponge species (1,232 specimens) without a priori knowledge. These sponges were grouped in four clusters, from which the largest two were composed of species consistently predicted as HMA (n = 44) and LMA (n = 74). In summary, our analyses shown distinct features of the microbial communities associated with HMA and LMA sponges. The prediction of the HMA-LMA status based on the microbiome profiles of sponges demonstrates the application of machine learning to explore patterns of host-associated microbial communities.

Highlights

Sponges (Porifera) represent one of the oldest, still extant animal phyla
The contradictory classification of E. formosus as being low microbial abundance” (LMA) (Gloeckner et al, 2014) or high microbial abundance” (HMA) (Easson and Thacker, 2014) was revisited and based on the present transmission electron microscopy (TEM) images documenting large amounts of microorganisms (Figure 1), E. formosus was clearly identified as an HMA sponge
As demonstrated for the HMA-LMA dichotomy, we have shown that such effects are observed at different taxonomic scales, e.g., when Operational Taxonomic Units (OTU) abundances are grouped at the phylum, class, or order level

Summary

Introduction

Sponges (Porifera) represent one of the oldest, still extant animal phyla. Fossil evidence dating back 600 million years ago shows their existence in the Precambrian (Yin et al, 2015) long before the radiation of all other animal phyla. Sponges are associated with microbial communities, with representatives of 41 different prokaryotic phyla far recovered from sponges, from which 13 phyla were shared among the 81 host species surveyed (Thomas et al, 2016). Sponge species were observed to harbor dense communities of symbiotic microorganisms in their tissues, while others were found essentially devoid of microorganisms (Reiswig, 1974). They were firstly termed “bacterial sponges” and “non-symbiont harboring, normal sponges” (Reiswig, 1981) and later the terms high microbial abundance” (HMA) and “low microbial abundance” (LMA) were used (Hentschel et al, 2003). Considering that more than 8,500 formally described sponge species exist and that the true diversity is still much higher (van Soest et al, 2012), a comprehensive survey of the HMA-LMA pattern would be a difficult and laborious undertaking

Methods

Results

Discussion

Conclusion