Generation and analysis of 280,000 human expressed sequence tags.

TLDR

The study aims to generate and publicly deposit ESTs from diverse human genes. ESTs were produced by high‑throughput single‑pass sequencing of 194,031 cDNA clones from 26 oligo(dT) libraries (18 normalized) derived from 17 tissues, with automated annotation and deposition. The EST dataset contains numerous known and novel sequences, and HMM analysis indicates that normalization reduces but does not eliminate gene‑family redundancy.

Abstract

We report the generation of 319,311 single-pass sequencing reactions (known as expressed sequence tags, or ESTs) obtained from the 5' and 3' ends of 194,031 human cDNA clones. Our goal has been to obtain tag sequences from many different genes and to deposit these in the publicly accessible Data Base for Expressed Sequence Tags. Highly efficient automatic screening of the data allows deposition of the annotated sequences without delay. Sequences have been generated from 26 oligo(dT) primed directionally cloned libraries, of which 18 were normalized. The libraries were constructed using mRNA isolated from 17 different tissues representing three developmental states. Comparisons of a subset of our data with nonredundant human mRNA and protein data bases show that the ESTs represent many known sequences and contain many that are novel. Analysis of protein families using Hidden Markov Models confirms this observation and supports the contention that although normalization reduces significantly the relative abundance of redundant cDNA clones, it does not result in the complete removal of members of gene families.