When it was proclaimed that the Library comprised all books, the first impression was one of extravagent joy. All felt themselves lords of a secret, intact treasure. There was no personal or universal problem whose eloquent solution did not exist... the universe suddenly expanded to the limitless dimensions of hope. --Borges
In the mythical library of Babel [19], Borges describes a combinatorially vast collection of books. Each individual book in the population is of a standard format, sharing size, number of pages, and characters per page. The body of their text consists of letters from a finite alphabet. Each book resides on a shelf beside others in a room among countless other rooms. It is easy to imagine that every possible book is present in the combinatorially vast collection.
In addition to texts, the library is populated by librarians, who studiously search through each text for hints of meaning. Meaning might appear in one of many forms: a book of books, which indexes or explains where one may find any answer to any question; a history of the library or an account of its development; a pathway of references from one book to another; answers to the mysteries that confound humanity. The librarians analyze the texts in search of sense. They attempt to understand their realm by scanning for a meaningful series of words on a page, or a book that stands out among its cohort as containing minimal gibberish, or simply a repeated pattern.
Paradoxically, a library of this magnitude does not allow for the possibility of an outside, an external realm to which the entity can refer, or with which it can interact. The system is closed; the library is a universe. Generations of librarians have known no other than endless shelves of texts of relentless strings of characters.
The status of biology in the post-genome era is in some ways not unlike the library at Babel. Living populations are limited from every possible or conceivable variant to the realm of the actual, as discovered by evolutionary dynamics. However, a combinatorial enormity exists among the many genomes in a population of a single species, or among all species, living or extinct. Though not literally residing on shelves in some physical library, the information content of many genomes is akin to the texts in Babel's collection, equally useless without an index, or a set of organizing principles.
As the universe of books in Babel's library makes no sense without the context provided by an outside world, genes in model organisms, and even model organisms as entities, do not make sense without considering the contexts in which they interact. Understanding the role of a particular gene or protein requires knowing the greater context in which it functions [10,17]. The situation of studying species in a larger ecological context has been described as the coevolutionary perspective. Summarily, ``Species in pure isolation simply do not make sense'' [116].
Lest functional genomic research be cloistered in the realm of solipsism, considering these more general contexts is crucial to understanding how organisms function and respond to their environment. Knowing the sequence of a gene does not account for its function, as knowing the sequence of a genome is not sufficient to account for the dynamic phenomenon of life. Both are preliminary.
A central mystery of life is how the diversity and complexity that surround us emerged from the laws of physics, chemistry, Darwinian evolution, and Mendelian inheritance. A counterbalance to the detailed, yet static view of the genome perspective is offered by a dynamical systems perspective. Life is the antithesis of stasis. Indeed, it is the dynamic interactions among the many different parts of cells, organisms, communities, and ecosystems that result in the phenomenon of life. The dual challenge of contemporary biology is to know both the individual components of a living system and how they interact.
This work has focussed largely on attributes of large sets of transcribed sequences. In effect, the methods described in this dissertation provide several ways to partition sets of sequences obtained from symbiotic interactions. The methods can be used to divide a large set of sequences between the species of origin, between those transcripts sequenced and those yet unsequenced, or between those transcripts distinct to one library and those common to others. We have not yet identified specific molecular mechanisms of symbiosis, but this is not the exclusive role of computational biology. My intent has been to facilitate discovering mechanisms of mutualism by advancing ways to partition the seas of data that result from high-throughput sequencing of expressed tags.
I have taken a rather static approach to discovering mechanisms of mutualism. Considering the dynamics of cellular processes as a network or a graph, and the interactions between species as a consequence of two interacting, but interdependent, networks, most effort has been devoted to identifying nodes in the respective graphs, rather than the connections between and fluxes among those nodes. This is necessary in early stages of investigation, but not sufficient to understand fully the dynamic nature of the interactions between symbionts. Even as we aspire to ascend a mountain peak, we cannot reach the summit without first navigating difficult terrain. With future investigations and technical innovations yet unforeseen, one hopes to ascend beyond the current state of understanding. In this sense, these are my first steps.