NCBI Magic-BLAST RNA-seq mapping tool

• Home
• Documentation
• Feedback
• Download
• How Magic-BLAST works
• Output
• Cookbook
• Access SRA reads in the cloud
• Use NCBI SRA repository
• RNA vs DNA
• Paired reads
• Multi-threading
• Reads in FASTA or FASTQ
• Create a BLAST database
• Developers
• License
• Exceptions
• Copyright
• Release Notes
• Release Notes

Splicing

By default, Magic-BLAST aligns RNA reads to a genome and reports spliced alignments, possibly spanning several exons. To disable spliced alignments, use the -splice F option.

For example, mapping RNA or DNA reads to a bacterial genome:

magicblast -sra SRR5647973 -db salmonella_enterica_genome -splice F

Transcriptome

Use the -reftype transcriptome option, to map reads to a transcriptome database. For example:

magicblast -query reads.fa -db my_transcripts -reftype transcriptome

The -ref_type transcriptome option is a short hand for -splice F -limit_lookup F, so the above call is equivalent to:

magicblast -query reads.fa -db my_transcripts -splice F -limit_lookup F

Magic-Blast finds alignments between a read and a genome based on initial common word in both. Many genomes contain interspersed repeats that make mapping much more time consuming. To make mapping faster we disregard words that appear too often in the reference. This is not desirable when mapping to transcripts, because a transcript with many variants could be considered a repeat. The -limit_lookup F option turns this functionality off.