New Technologies in Human Genetic Analysis by Mark E. Samuels, Brent Higgins, Sylvie Provost, Julien Marcadier, Christian Blouin, Sharen Bowman, and Marie-Pierre Dubé G E N O M I C / P R O T E O M I C T E C H N O L O G Y Human genetics has benefited in the last decade from spectacular technological advances in DNA analysis. The rate of data acquisition has increased by orders of magnitude. To cope with this flood of information, semiautomated methods have been developed for primary data collection and interpre- tation of genotyping and direct DNA sequencing. Here the authors examine several improvements in experimental design and analysis, expediting disease gene discovery in family-based genetic studies. Microsatellite genotyping Microsatellite markers, also known as short tandem repeats (STRs), arise from consecutive repeat units such as CACACA . . . or GATAGATAGATA . . . embedded within unique sequences. 1 The human genome contains many thousands of such repeats of di-, tri-, and tetranucleotide types. 2 PCR primers designed in the surrounding unique sequence allow unambiguous amplification of particular microsatel- lites. Many of these sites are polymorphic within the human population, segregating multiple alleles with different repeat lengths. These PCR amplicons can serve as genetic markers for the corresponding chro- mosomal segments as they segregate through families. Microsatellite marker length is normally conserved over several meiotic generations, making these effec- tive tools for pedigree-based linkage analysis. 3 Microsatellites offer several technical challenges. 4 PCR enzymes tend to stutter at the repeats, leading to a series of smaller-than-full-length peaks whose lengths usually differ by one repeat unit. 5 More serious is the tendency of these enzymes to add an additional, nontemplated nucleotide at the 3′ end of the product. When the extent of such addition is variable, this leads to the phenomenon of peak splitting, which can render dinucleotide markers practically useless. One way to reduce variability in peak splitting is to add a specific sequence tag at the 5′ end of one of the PCR amplification primers. When fluorescent genotyping is being performed, this should be the nonlabeled primer (often though not always formalized as the reverse primer). Several such sequence tags have been reported, 6,7 although their exact mechanism of action remains uncertain. The authors have verified the effectiveness of such a sequence tag, using either specialized plusA or nomi- nally standard PCR conditions. As shown in Figure 1a and b, two different dinucleotide markers varied in the sensitivity of splitting to the PCR protocol employed when default primer design was used. However, addi- tion of the 5′ tag on the unlabeled primer minimized variability for both markers, making genotype calling less problematic in each case. This suggests that the tag should routinely be included on all customized microsatellite markers as a preventive measure. SNP genotyping Single-nucleotide polymorphisms (SNPs) pro- vide an alternative to microsatellites for genetic mapping. Individual SNPs are overwhelmingly biallelic, hence intrinsically less informative than microsatellites. However, multiple SNPs analyzed in tandem have the potential to carry the same or greater information content than microsatellites. Several millions of SNPs have been genotyped as part of the HapMap project, providing allele frequencies in several different populations. 8 High-density SNP panels are now commercially available. When used appropri- ately, these panels can be used for genetic map- ping of monogenic disorders in the traditional family-based paradigm. 00 / FUTURE 2007 • AMERICAN BIOTECHNOLOGY LABORATORY To compare the effectiveness of dense SNP panels to microsatellite markers for pedigree-based genetic mapping, the authors sought to verify a known genetic linkage, using a large Nova Scotia Acadian family segregating the recessive trait Niemann-Pick type D. 9,10 The underlying causal gene was identified as NPC1 (OMIM #257220) by positional cloning, 11 and homozygous mutations in NPC1 were confirmed in affected individuals in the Nova Scotia kindred. 12 The authors genotyped two distantly related affected individuals plus one unaffected sibling from this fam- ily (see Figure 2) using the Xba 50K chip (Affyme- trix, Santa Clara, CA). As seen in Table 1, the lon- Figure 1 Microsatellite genotyping before/after reverse 5′ sequence tag. a) Fluorescent primers for D2S133 were used to amplify DNA from one randomly selected sample, which is homozygous for a particular allele of this marker. Amplification was performed under two different PCR protocols, using either the tagged or untagged reverse primer. From top to bottom: minus tag, plusA PCR; minus tag, standard PCR; plus tag, plusA PCR; plus tag, standard PCR. In the absence of the reverse tag, use of plusA PCR conditions biases products toward nontemplated nucleotide addition, as seen by comparing the first and second panels. The addition of the reverse tag increases amplicon size, as shown in the two bottom panels, and reduces variability in peak splitting under different PCR conditions. Note that the stutter peaks at 2-bp intervals are independent of PCR conditions and presence of reverse tag. GeneMarker genotyping software was used to identify and call the allele as 297 bp in the bottom panels (gray vertical bar). b) D16S520, from top to bottom: minus tag, plusA PCR; minus tag, standard PCR; plus tag, plusA PCR; plus tag, standard PCR. Similar to panel (a), but this marker is heterozygous for alleles at 196 and 198 bp, as called by GeneMarker (gray vertical bars) for the reverse tagged version of the marker. Peak splitting is much more severe for this marker, but addition of the reverse tag still reduces the variation significantly. As above, addi- tion of the reverse tag increases the size of the amplicon. a b