One motive underlying this work has been to uncover mechanisms of mutualism by advancing methods to analyze sequence data from diverse sources. Such data must be refined in various ways before patterns emerge from the background noise that always exists in experiments that result from uncoordinated and partial efforts.
What have we learned? By summarizing the partial state of understanding the mechanisms by which AM fungi grow, develop, and establish healthy mutualisms with plant roots, we have focused our attention on processes essential to an ancient and important mutualism. Continued application of molecular and genetic techniques will generate large volumes of data. Careful computational analyses might greatly enhance our perspective. The extra advantage of examining properties of several libraries at once allowed us to gain insights that would not have been obtained from looking at individual genes, or even a collection of transcripts within a single library. For instance, our analyses detected fungal transcripts from three libraries that resemble plant gene sequences in terms of hexamer composition, but not sequence similarity.
Our attention is much attracted to the prevalance of plant-like transcripts in fungal libraries and fungal transcripts in plant libraries. Preliminary follow-up analysis results, using specific detection of ubiquitin of either plant or fungal origin in plant and fungal libraries (based on [103], but using BLAST similarity searches), suggest that some cross-species contamination may be present in reportedly pure libraries. This could explain some, but perhaps not all, such examples. The availability of replicate or independent experiments would help to resolve this issue. That it may be an issue was only established by the work described here.