DeCypher Algorithms

Products

Applications

News

Support

Sales

Contact

accelerated sequence comparison solutions

Section Topics:

• Introduction
• SeqCruncher Accelerator
• Tera-BLAST
• Tera-Probe
• GeneDetective
• HMM Solution
• Smith-Waterman
• Algorithms
• Interfaces
• Published Citations

News:

Active Motif is excited about the upcoming PAG 2010 meeting in San Diego.

8 ways CodeQuest™ can improve your 454 data analysis

Active Motif is exploring the use of FPGAs for Next-Generation Sequencing Data Analysis

Comparisons with HMMER and SSEARCH show that DeCypher delivers 300-1000X improved performance over CPU core.

Newsletter:

Subscribe to the TimeLogic newsletter for product information, special offers and technology whitepapers!

Informatics Applications available for DeCypher and CodeQuest

Tera-BLAST™
Algorithm	Query	Target	Description	Applications
Tera-Blast™ BLASTN	NT	NT	Fast NT search, find very similar matches, ungapped search	map EST or cDNAs onto genome find vector contaminations find overlaps for assemblies genome vs. genome comparison
Tera-Probe™	NT	NT	High-speed, high-sensitivity NT search for large sets of oligo-sized queries. Tolerates multiple mismatches	map oligos to a genome for microarray design, primer design, and siRNA studies
Tera-Blast™ BLASTP	AA	AA	Fast AA search, find easy homologous sequences, ungapped search	assign tentative functions homology modelling when a match to protein with known 3D structure is found
Tera-Blast™ BLASTX	NT	AA	Translate query in all 6 frames, search in AA space., ungapped search	assign tentative functions to novel cDNAs or EST
Tera-Blast™ TBLASTN	AA	NT	Translate target in all 6 frames search resulting AA sequences, ungapped search	map proteins onto genome
Tera-Blast™ TBLASTX	NT	NT	Translate query and target in all 6 frames, search the AA sequences, ungapped search	compare genomes, more sensitive for comparing coding regions
GeneDetective™
Algorithm	Query	Target	Description	Applications
GeneDetective	cDNAs or ESTs genomic	genomic cDNAs or ESTs	Utilizes Tera-BLAST and full Smith-Waterman alignments to map ESTs to genomic DNA	gene modeling visual splice variant analysis
	Protein genomic	genomic protein		gene modeling visual splice variant analysis
	Protein HMM	genomic	Maps hidden Markov Models onto genomics sequences to define gene model. Uses proprietary intron-spanning HMM-Framesearch to tolerate low quality sequences	gene modeling visual splice variant analysis
	genomic	Protein HMM		gene modeling visual splice variant analysis
DeCypherSW™
Algorithm	Query	Target	Description	Applications
Smith-Waterman	NT	NT	like BLASTN, more sensitive, allow gaps	find and mask repetitive elements find regulatory regions
	AA	AA	like BLASTP, more sensitive, allow gaps
	NT	AA	like BLASTX, more sensitive, allow gaps
Double-Affine Smith-Waterman	NT	NT	SW, allow very long gaps	map cDNAs onto genomes, can gap over long introns
	AA	AA	SW, allow very long gaps	situations where entire domains are deleted between otherwise homologous proteins
	NT	AA	SW, allow very long gaps	map proteins onto genomes, can gap over long introns, if in frame
FrameSearch	NT	AA	like BLASTX, more sensitive, can align across gaps and frameshifts	assign tentative functions to EST
FrameSearch	AA	NT	like TBLASTN, more sensitive, can align across gaps and frameshifts	assign tentative functions to EST
Symmetric Frame Independent™	NT	NT	Solves indel issues with EST and other low coverage nucleic data by conducting a sensitive protein domain search between your nucleic query and the nucleic database, with frame shifting as needed.	like TBLASTX, more sensitive, allows for gaps and frameshifts
DeCypherHMM™
Algorithm	Query	Target	Description	Applications
Hidden Markov Model	NT	HMM	Translate query in all 6 frames, search HMM database.	assign protein families, domains and functions to cDNAs or genomic sequences
	NT	NT HMM	Compare NT sequence to NT HMM database.	find repetitive elements find regulatory regions: e.g. transcription factor binding sites
	AA	HMM	AA sequence vs. HMM database.	assign domains and functions to protein sequences predict 3D structures
	HMM	AA	Compare HMM to AA sequence database.	find new members ot the family of proteins in HMM train HMM
HMM FrameSearch	NT	HMM	Translate query in all 6 frames, search HMM database, allow frameshifts in query	assign protein families, domains and functions to cDNAs or genomic sequences
HMM FrameSearch	HMM	NT	Translate target in all 6 frames, search with HMM, allow frameshifts in target
Profilescan	NT	Profile	Translate query in all 6 frames, search profile database.	assign protein families, domains and functions to cDNAs or genomic sequences
	NT	NT Profile	Compare NT sequence to NT profile database.	find repetitive elements find regulatory regions: e.g. transcription factor binding sites
	AA	Profile	AA sequence vs. profile database.	assign domains and functions to protein sequences predict 3D structures
Profilesearch	Profile	AA	Compares profile to AA sequence database.	find new members ot the family of proteins in profile train profiles
Profile FrameSearch	NT	Profile	Translate query in all 6 frames, search profile database, allow frameshifts in query	assign protein families, domains and functions to cDNAs or genomic sequences
Profile FrameSearch	Profile	NT	Translate target in all 6 frames, search with profile, allow frameshifts in target	assign protein families, domains and functions to cDNAs or genomic sequences

Non-Accelerated Algorithms

ClustalW, with graphical rendition of dendrogram via Java applet

Algorithm Development

TimeLogic adapts established algorithms to run on our FPGA Accelerator Arrays by reinterpreting from first principles. Contact us today for additional information on DeCypher's accelerated algorithm services.