MOTIVATION: Discovery of novel splicing from RNA sequence data remains a critical and exciting focus of transcriptomics, but reduced alignment power impedes expression quantification of novel splice junctions. RESULTS: Here, we profile performance characteristics of two-pass alignment, which separates splice junction discovery from quantification. Per sample, across a variety of transcriptome sequencing datasets, two-pass alignment improved quantification of at least 94% of simulated novel splice junctions, and provided as much as 1.7-fold deeper median read depth over those splice junctions. We further demonstrate that two-pass alignment works by increasing alignment of reads to splice junctions by short lengths, and that potential alignment errors are readily identifiable by simple classification. Taken together, two-pass alignment promises to advance quantification and discovery of novel splicing events. CONTACT: [email protected], [email protected] AVAILABILITY AND IMPLEMENTATION: Two-pass alignment was implemented here as sequential alignment, genome indexing, and re-alignment steps with STAR. Full parameters are provided in Supplementary Table 2. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Despite significant advancement in alignment algorithms, the exponential growth of nucleotide sequencing throughput threatens to outpace bioinformatic analysis. Computation may become the bottleneck of genome analysis if growing alignment costs are not mitigated by further improvement in algorithms. Much gain has been gleaned from indexing and compressing alignment databases, but many widely used alignment tools process input reads sequentially and are oblivious to any underlying redundancy in the reads themselves. RESULTS: Here we present Oculus, a software package that attaches to standard aligners and exploits read redundancy by performing streaming compression, alignment, and decompression of input sequences. This nearly lossless process (> 99.9%) led to alignment speedups of up to 270% across a variety of data sets, while requiring a modest amount of memory. We expect that streaming read compressors such as Oculus could become a standard addition to existing RNA-Seq and ChIP-Seq alignment pipelines, and potentially other applications in the future as throughput increases. CONCLUSIONS: Oculus efficiently condenses redundant input reads and wraps existing aligners to provide nearly identical SAM output in a fraction of the aligner runtime. It includes a number of useful features, such as tunable performance and fidelity options, compatibility with FASTA or FASTQ files, and adherence to the SAM format. The platform-independent C++ source code is freely available online, at http://code.google.com/p/oculus-bio.