Introduction | Description | Tests | Summary | References | FAQ | Program Interface

INTRODUCTION

Comparative studies of genes and genomes are becoming increasingly common in biology, and include studies of molecular evolution (1), organismal evolution (2), and genetic mapping of functionally characterized genes (3, 4). Most such studies rely on the polymerase chain reaction to amplify orthologous genes among related organisms, which creates a need for efficient methods to design primers from a nucleotide alignment. For many such studies, the organisms are similar enough that DNA sequences of genes and often even non-coding sequences can be aligned, but dissimilar enough that PCR primers designed from a single species fail to bind efficiently. The taxonomic range of such groups varies depending on the gene or genes, but examples include cereal grasses (5, 6), Brassicaceae (7), mammals (8, 9), birds (10), and families or genera within those groups. For labs such as ours that are investigating multiple species, PCR primers need to be designed to work reliably and specifically on dozens of species without additional optimization. We find that the ideal primers for this are minimally degenerate. A similar challenge is faced by labs that map multiple genes in two or more genomes.


Design of non-degenerate or minimally degenerate PCR primers is often done manually, and is time-consuming, often taking most of a day. One starts with an alignment of two or three sequences of related species, usually downloaded from GenBank. These are then scanned for two regions of sequence identity separated by an appropriate number of base pairs, and the regions then used as input into a primer design program to check melting temperatures and presence or absence of predicted secondary structures. For study of a single gene this is easy, but for high-throughput analysis of multiple genes, the time investment is prohibitive. The difficulty in designing primers for phylogenetic studies is a proven bottleneck and has resulted in relationships among species being based on a comparative handful of markers, despite extensive evidence that some of these markers have serious limitations (2, 11).


Primer design software is available for some applications. For designing primers for a single nucleotide sequence, web-based and stand-alone software packages are available, such as Primer3 (12) and NetPrimer (Premier Biosoft International). For designing primers from alignments of highly divergent proteins, Codehop (13) uses a mix of consensus bases and degeneracy.


There remains the problem of designing non-degenerate or moderately degenerate primers to alignable sequences of nucleotides, whether coding or not. Accordingly, we have created a free, web-based, multiple species, primer prediction application, which we call Primaclade.


Citaton for Primaclade: M. D. Gadberry, S. T. Malcomber, A. N. Doust, and E. A. Kellogg. 2005. Primaclade - a flexible tool to find primers across multiple species. Bioinformatics 21: 1263-1264.


Introduction | Description | Tests | Summary | References | FAQ | Program Interface