I asked ChatGPT this question one evening and working through the session was fun. Maybe it needs more fact-checking…I know some of my readers are fastidious enough to call me out if I’m wrong but here goes.

When I started in Genomics it was still called DNA sequencing and Northern Blots. I ran the sequencer in the Department of Pathology for Prof John Trowsdale and we bought a shiny new ABI377 (there are none on eBay for Nava to pull apart) when we moved the lab from the old ICRF building in Lincoln’s Inn. If my memory serves me correct the best I got to was almost 1000bp on the 96 lane stretch mode: a hardware and software configuration, which used longer gels and optimized electrophoresis conditions to get to ~1000 bp per sample under optimal conditions. This was probably the highest-throughput acrylamide gel-based DNA sequencing system that ever lived.

But all that effort – pouring those gels was like living in a bubble filled nightmare – and at best we could sequence 96kb per run!

Compared to today’s Illumina NovaSeq systems, which yield terabases per run, those acrylamide gel systems were labor-intensive, slow (overnight runs), and required manual cleanup and handling (aargh, I forgot about the washing up).

How Many Lanes to Match a NovaSeq X Plus Run?

I asked ChatGPT to do the math to compare the ABI 377 Stretch with an Illumina NovaSeq X Plus and work out how many 377 Stretch gels would be required…and it’s a lot!

  • NovaSeq per run (2 × 150 bp): up to 26 billion reads or ~7.8 trillion base pairs per run (depending on cluster density and paired-end read length)
  • ABI 377 Stretch Mode per run: 96 1000bp reads or 96 kilo base pairs per run.

That’s some big ABI 377xl glass plates!

It turns out you would need over 81 million ABI 377xl Stretch gels to match one NovaSeq X Plus run. And at maybe 2 gels a day (I can’t remeber if that was actually possible) that would equate to over 100,000 years of non-stop gel runs.

Keeping this crazy exercise going even further I asked ChatGPT to figure out just how absurdly massive this would be in real-world terms.

A standard ABI 377 Stretch gel was about 48 cm (height) × 38 cm (width) or approximately 0.1824m2. So 81.25 million gels would total around 14km2! Equivalent to over 2000 football pitches, 7x the area of Monaco or 3x the area of the Vatican City (ChatGPT got creative).

If you liked this then check out my how many MinIONs could you fit in a PacBio (might need updating to an Utlima 1G as PacBio shrunk)!