A rapid method for determining clone frequency in complex populations using PCR and the Poisson distribution

There are instances in which it is desirable to know the frequency of a particular cloned nucleic acid species in a complex population. For instance, before undertaking isolation of a clone from a library, it is helpful to know how many clones should be screened; also, the frequency of a clone in a cDNA population may be taken as a first approximation of mRNA frequency (keeping in mind possible cloning bias). In our case we are interested in determining the frequency of specific clones before and after cDNA enrichment experiments. The typical way to score clone frequency is by filter hybridization using nucleic acid probes. This is laborious and time consuming, and is often complicated by artifactual spots on the filters. Bloem and Yu (1) described a way to use the polymerase chain reaction (PCR) to decrease the complexity of a cloned population prior to screening by filter hybridization. After eluting phage heads from a filter replica of plated phage, PCR was performed to determine if the clone of interest was in the population on that plate; further subdividing a positive plate and repeating the analysis narrowed the search for the clone by several orders of magnitude. This suggested to us a simple way to score clone frequency using PCR. By diluting an E. coli library to a complexity such that the presence of a clone of interest falls within the range of the Poisson distribution (2), a simple plus/minus test can be applied to replicate growths subjected to PCR with appropriate primers. The null term of the Poisson distribution and the library titer are used to calculate the frequency of the clone of interest. The method will be described for libraries in E. coli cells, but there should be several ways to apply it to phage libraries, most simply by filter elutions as described by Bloem and Yu, or by phage growths in liquid culture.