Abstract LB-97: Optimizing SELEX with high-throughput sequencing

Abstract SELEX (systematic evolution of ligands by exponential enrichment) identifies oligonucleotide “aptamers” that bind their protein targets with high affinity and specificity. Beyond inhibiting all protein function, aptamers can target specific functional regions (e.g., phosphorylation sites and catalytic domains) leaving other function intact. This selective inhibition is a valuable tool for dissection of signal transduction pathways, such as the growing list implicated in cancer by sequencing efforts interrogating tumor genomes. New technological innovations in aptamer discovery aim to keep pace with this need by improving the efficiency of the traditional SELEX enrichment process wherein oligonucleotides are iteratively bound to target, eluted, and amplified to purify an initial random library to a few high-affinity aptamers. Principal among these advances is the adoption of high-throughput sequencing (HTS-SELEX) to detect enrichment of candidate aptamers within the evolving pool. The improved sensitivity of high-throughput sequencing over the traditional use of cloning and sequencing reduces the number of rounds required to reach detectable enrichment. As with traditional SELEX, enrichment and hence number of rounds and overall efficiency are dependent on the free target and oligonucleotide pool concentrations. Though early theoretical approaches have addressed these issues for traditional SELEX, realizing the potential efficiency gains of the new HTS-SELEX approach requires an analysis and resulting protocol tailored to it specifically. Here we provide such an analysis and also re-examine earlier theoretical approaches to traditional SELEX. We show that HTS-SELEX motivates an optimization condition different from that of traditional SELEX and consider the role of stochastic fluctuations in the relatively low multiplicities of the enriched species resulting from the earlier enrichment detection of HTS-SELEX. Unlike previous approaches seeking to maximize some measure of per-round enrichment, this analysis is a global optimization of the entire SELEX process. It results in a specific recommendation for free target concentration, constant across rounds and simply related to the desired affinity, and for corresponding total ligand concentration. Applying the global analysis to an optimization condition appropriate for traditional SELEX again yields recommendations that are provably optimal and that differ from those previously published. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr LB-97. doi:1538-7445.AM2012-LB-97