Tera-Probe™ for sensitive, high-throughput oligonucleotide searches

Tera-Probe™ (pdf brochure) combines the performance of Tera-BLAST¹ with efficient Smith-Waterman alignment methods to identify oligonucleotide sequences across a genome. Tera-Probe exceeds NCBI BLAST² sensitivity and matches Smith-Waterman³ results, yet offers much greater throughput.

Both DeCypher® and CodeQuest™ biocomputing systems process Tera-Probe faster than sixteen thousand individual CPUs running Smith-Waterman, making these solutions ideal for research teams designing microarray probes, mapping SNP databases and designing RNA interference experiments

Selecting Optimal Microarray Probe Sequences

Comparing short sequences to a genome database is essential for minimizing non-specific hybridization in microarray probe design. Tera-Probe offer significant advantages compared to other oligo search methods.

Smith-Waterman

• Often used to screen potential probe sequences, as it comprehensively explores all possible pairwise alignments to select an optimal local alignment.
• Smith-Waterman software (ssearch) is slow even on fast CPUs, with processing limited to 20 million cell comparisons per second per CPU.

NCBI BLAST

• Uses heuristic shortcuts to focus on anchor regions which share some core similarity, resulting in faster searches at the expense of sensitivity.
• May miss hits when multiple short gaps or mismatches occur between anchor regions

Tera-Probe

• Processes oligonucleotide search pipelines at 330 billion cell comparisons per second, or 16,500x faster than Smith-Waterman on 1 CPU.
• Requires only the space, power and cooling of a single workstation or server
• Can be scripted into your analysis pipelines
• Forces the entire oligo query to align to local target loci to ensure that all potential hybridization sites are examined
• Matches the sensitivity of Smith-Waterman software
• You can employ a custom scoring matrix to establish finely tuned scoring for matches and mismatch penalties
• Available with DeCypher Tera-BLAST and included with the CodeQuest workstation

Figure 1. Smith-Waterman (ssearch) was verified on a single CPU to compare 20 million cells per second. A single DeCypher Engine system compares 330 billion cells per second, for 16,500x the performance.

Figure 2. Comparing 10,000 oligonucleotide sequences (30-mers) to Unigene Mouse using DeCypher and GeneMatcher

A DeCypher system with 4 cards and 4 CPUs compares 10,000 oligos to Unigene mouse with Tera-BLAST in 1.5 hours. This requires 6 hours on a Paracel GeneMatcher2 system with 9 boards and 9 CPUs. Due to improved gap handling at the ends of each alignment, Tera-Probe identifies an additional 9,000 unique hits missed by GeneMatcher.

References

1. Luethy et al. Hardware and software systems for accelerating common bioinformatics sequence analysis algorithms. Biosilico 2(1), 2004
2. Altschul, SF, et al. Basic local alignment search tool. J Mol Biol 215(3), 1990.
3. Pearson, WR. Effective protein sequence comparison. Methods Enzymol. 266, 1996.

Contact a TimeLogic account executive today to learn more about Tera-Probe today!

Download Complete Tera-Probe brochure