Adam Savage’s Tested
2 HN recommenders · recommended in 3 months, 2020–2024
1:11:00 · 52K views · Oct 31, 2018
An episode of Adam Savage's Still Untitled podcast with molecular biologist Joe DeRisi. They discuss his work, including sequencing to detect rare infections, in a conversation Savage calls possibly the show's scariest.
HN recommends it for DeRisi's fascinating stories about rare pathogens such as brain-eating amoebas.
3 comments from Hacker News
Adam Savage's podcast Still Untitled did a really fantastic episode [1] back around Halloween 2018 where they had a molecular biologist come on and talk about his work. The stories he tells are absolutely fascinating and I thoroughly encourage everyone wiht any interest in molecular biology and crazy stories to check it out. I won't go into it because I'll spoil it, but suffice it to say that normal episodes are 20-30 mins, this one went for 70 mins and the first time I listened to it I immediately restarted it to listen to it again. The first half of the episode is a story that gets more and more insane as it unfolds. Afterwards, he goes into a few other shorter stories of his work. Anyway, the link to this is because there's one story from about 38:15 to about 45 minutes in that discusses brain-eating amoebas he was involved in. It's not this exact one, but it's related to it. I recommend anyone who finds this interesting to check it out. Although there's video on youtube, you really don't need video for it and your podcast app of choice should have an audio version of it. [1]: this video
From what I understand Joe DeRisi's approach[0] comes relatively close to your "blender" idea! [0]: I already mentioned Adam Savage's excellent podcast with him elsewhere in this dicussion but here's a direct link: this video
Re step 1 and 2 - here's interesting podcast on how they detect rare infections: this video Basically sequence everything what's in your blood and look for what doesn't match your genome === infection. The problem is this is orders of magnitude more compute intense than whole genome sequencing. Basically increased demand for sequencing far outmatches available compute!