Germany, Europe. The author of the genome-assembly software MIRA, Bastien Cheval, has finally admitted that he does all of the software’s assembly tasks in his head.
MIRA, deemed the “swiss-army knife of genome assembly”, has been around for over 15 years and many use it as their “weapon of choice” for complex genome assembly projects, especially those that involve more than one technology (Illumina, 454, PacBio etc).
However, suspicions were raised recently when a group of scientists calling themselves “Assemblathong 4000” delved into the code to find nothing other than an e-mail program set up to e-mail sequence data directly to the author, Bastien Cheval. There was no core algorithm. Assembly was not happening within the software.
Initial fears that this was an attempt to steal data were quickly quashed when it became clear that that would be a really, really stupid way of doing it.
When confronted with the software anomaly, Bastien had this to say: “Look, I spent months writing some assembly algorithms, but in the end I found I could just do it quicker in my head. I’m like The Rain Man of genome assembly; I see the sequences and I can just assemble them in my head. I think my brain is a string graph. It just happens. E-mailing data to me seemed the quickest way of completing my user’s assembly tasks”
After initially expressing shock and awe, users are now used to the idea, and feel comfortable using MIRA in this way. Attempts are under way to clone Bastien to provide a parallel version of MIRA.