PHACCS, an online tool for estimating the structure and diversity of uncultured viral communities using metagenomic information

Abstract

Background: Phages, viruses that infect prokaryotes, are the most abundant microbes in the world. A major limitation to studying these viruses is the difficulty of cultivating the appropriate prokaryotic hosts. One way around this limitation is to directly clone and sequence shotgun libraries of uncultured viral communities (i.e., metagenomic analyses). PHACCS http://phage.sdsu.edu/phaccs, Phage Communities from Contig Spectrum, is an online bioinformatic tool to assess the biodiversity of uncultured viral communities. PHACCS uses the contig spectrum from shotgun DNA sequence assemblies to mathematically model the structure of viral communities and make predictions about diversity.

Results: PHACCS builds models of possible community structure using a modified Lander-Waterman algorithm to predict the underlying contig spectrum. PHACCS finds the most appropriate structure model by optimizing the model parameters until the predicted contig spectrum is as close as possible to the experimental one. This model is the basis for making estimates of uncultured viral community richness, evenness, diversity index and abundance of the most abundant genotype.

Conclusion: PHACCS analysis of four different environmental phage communities suggests that the power law is an important rank-abundance form to describe uncultured viral community structure. The estimates support the fact that the four phage communities were extremely diverse and that phage community biodiversity and structure may be correlated with that of their hosts.

Figure 1

Flowchart of PHACCS. *The rank-abundance functions and the range of genotypes to use can be defined by the user. **This parameter represents b for the power law, logarithmic, lognormal and exponential distributions and k for the niche preemption. This parameter is not applicable to the broken stick.

Figure 2

Comparison of the structure and diversity of the different viral communities using PHACCS. The graphics represent rank-abundance curves, where the abundance of each genotype is plotted versus its abundance rank, the genotype of rank one being the most abundant. The curves were obtained by plotting the PHACCS rank-abundance values of the different communities on the same axis. *The predicted community structure for MBSED was the same for the lognormal, logarithmic, power and exponential rank-abundance forms. As a consequence, the diversity predictions were also the same.

Figure 3

Screenshot of PHACCS' advanced web interface.