Every day is exciting at Cofactor Genomics. Since we posted the new website, we have several pressing announcements!

Our Classroom Project winners have been announced. We are funding 4 undergraduate genomics projects, and some of those even more than the promised 1-lane/1-barcode level. These projects will serve both as experiments in molecular biology and as experiments in education. It is our hope that the syllabi, data, and methods, and perhaps even student testimonials, will serve as templates for undergraduate classes across the nation. Experiential learning is the best way to teach science, especially genomics, and we are proud to contribute to the great work of educators like Sarah “Sally” Elgin and the original Genomics Education Partnership. Several of us at Cofactor either assisted or took Sally’s class, so we encourage you to check it out.

We also launched the long-anticipated Cofactor Video! Bio-Rad and The PCR Song started the fad with Music Videos recast for science geeks. Eppendorf followed suite with a Boy-Band video about the epMotion. Somebody even remade a classic film into The AffyMatrix. Our innovation? Bring science geek-ery to a new genre – the Mock-u-mentary. Enjoy! Force your friends to watch it – that gel extraction can wait.

Last but not least, we can now generate 1 Giga-Base of pass-filter per lane per end on the Illumina using only 60bp reads. If you really wanted 106bp reads, sure then we could give you 1000MB / 60bp * 106bp = 1.8GB per lane per end, but those extra bases aren’t high quality enough to warrant the extra time. Certainly not if you want to put them into something like Velvet where adding more data that is poor quality actually makes the assembly worse. And if you’re not doing de novo assembly, we posit that 60bp is long enough!

Buyer beware! Why do we have this ridiculous expression “pass-filter per lane per end @ 60bp reads” ? These 4 variables all need to be held constant when comparing runs. Many places artificially inflate output amounts by any or all of these methods:

• quoting raw data output instead of filtered (inflates 50-30%)
• quoting output at 106bp reads which are garbage after the ~75th base (inflates 76%)
• quoting output for paired-end lanes when you are thinking single-end (inflates 100%)
• quoting output for an entire flowcell instead of lane-by-lane (inflates 800%)

Of course, when we talk about the SOLiD you have to take it one step further: “mappable per barcode per end per slide @ 50bp reads.” The SOLiD can run 2 slides at once without really increasing the run time. Since it runs twice a long as the Illumina this conveniently puts it back on par time-wise. Of course, you still get 50% more reads on the SOLiD.

We are always here to cut through the hype of each platform and optimally allocate projects across platforms. Let us know if you have any questions about Next-Gen!