With the rise in prominence of biological research using new short-read DNA sequencing technologies comes the need for new techniques for aligning and assigning these reads to their genomic location of origin. Until now, methods for allocating reads which align with equal or similar fidelity to multiple genomic locations have not been model-based, and have tended to ignore potentially informative data. Here, we demonstrate that existing methods for assigning ambiguous reads can produce biased results. We then present new methods for allocating ambiguous reads to the genome, developed within a framework of statistical modeling, which show promise in alleviating these biases, both in simulated and real data.

Affymetrix SNP arrays are widely used for both genome-wide association and copy number variation (CNV) studies, both of which depend on accurate copy number estimation. However, depending on the method used to copy number estimation, distortions from the actual copy numbers can occur. Therefore, we demonstrate here several effects that can bias accurate copy number estimation, and we describe how some of these biases can be adjusted by existing methods, while others require further study.

The dissection of complex diseases is one of the greatest challenges of human genetics with important clinical and scientific applications. Traditionally, associations were sought between single genetic markers and disease. The availability of large scale SNP data makes it possible, for the first time, to study the predictive power of genotypes and haplotypes with respect to phenotype data. Here we present a novel method for predicting phenotype information from genotype data. The method is based on a support vector machine that employs new kernel functions for the similarity between genotypes or their underlying haplotypes. We demonstrate our approach on SNP data for the apolipoprotein gene cluster in baboons, predicting plasma lipid levels with significant success rates, and identifying associations that were not detected using extant approaches.

We review recent developments in spaced seed design for cross-species sequence alignment. We start with a brief overview of original ideas and early techniques, and then focus on more recent work on finding accurate (sensitive and specific) seeds for cross-species cDNA-to-genome alignment. These recent developments include methods and models for estimating seed specificity and determining sensitive and specific seeds, finding seeds that can be applied to a wide range of comparisons, and applying seed models to other computational biology areas, such as gene finding.