* Total sequences deposited: 21,499. Total sequences in clustering above: 19,218. Data shown above does not include 600 high quality sequences used in clustering but were from several libraries which were not sufficiently complex for extensive sequencing (E1, E2, E5, H2, H4, H5, and NF). Also not shown is 2,281 sequences of sufficient quality for deposition into Genbank but that contained long polyA sequences and/or had lower quality near the beginning of the sequence.

^ Library recombinants, in millions of theoretical independent recombinants. ND= not determined.

# calculated as follows: at least 30 randomly selected clones were PCR-amplified and insert size calculated. Estimate was based on average of all amplified inserts over total inserts measured (at least 24 were required for calculation).

N - The number of sequences in a library.

Self - The Self-novelty of a library. This is the number of clusters the library forms divided by the total number of sequences (N).

Novelty - The number of clusters that consist only of that library within this dataset, divided by the total number of sequences (N).

Singletons - The number of singletons in the overall clustering contributed by this library.

## TIGR= The Institute for Genome Research (http://www.tigr.org/). Total cluster numbers shown are not directly comparable to our internal analysis due to different set of sequences used in TIGR cluster analysis (21,369).

+ Numbers indicate sequence totals for each library that are in tentative clusters (TC) or in singletons.