Dr. Bloom recovered #SARSCoV2 sequences that are closer to bats that any early 2020 human viral samples despite the order to destroy all early viral samples /2
Dr. Bloom applied equally insightful institutional sleuthing and sequence matching and evolutionary analysis to find sequences missing from the Short Read Archive @NIH but then resurrected their digital ghosts from @googlecloud /3
despite requests (without explanation) for the deletion of those sequences from the Short Read Archive /4
These sequences are then placed in context relative to the other sequences previously published /5
Which leads Dr. Bloom to wonder about the missed scientific and public health opportunity and why the sequences were deleted /6
This paper will be used in classes for years to demonstrate the full toolkit of sequence and evolutionary computational biology and the agency of biologically sophisticated data scientists. Students read this paper!
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Weight loss the data nerd way: This post by @karpathy on "Biohacking Lite" karpathy.github.io/2020/06/11/bio… was pointed out by colleagues including @AndrewLBeam who noted how trim he looked on AI Day. It's interesting at many levels A) Physics/CS take on the controversies 1/
in weight-loss and the hard truths of thermodynamics B) how biology is understood by someone who was not educated/warped by a conventional biological education C) The personality type that allows implementation of a short feedback look on calorific I/O. Med students would 2/
benefit from reading/discussing this case history and post. It also reminds me of my short essay @WBURwbur.org/news/2017/06/1… where I point where the thinking of implementation of AI in medicine is shallow 3/
What if you or your doctor could accurately predict how long you had to live upon a new diagnosis of aggressive cancer (e.g. lung cancer or sarcoma)? I've faced that question both as a doctor and with a dearly beloved. On the one hand I know I would 1/ #ML#AI#mortality
do the utmost to beat the odds forward.com/scribe/470514/… and a gloomy prediction would be just another hurdle to overcome. On the other hand, a very accurate predictor of mortality upon diagnosis would be very useful: we might dispense w control arms in trials [at our peril] 2/
It's not just that they realized that the new variants were being coded by DNA with unusual intragenic splices and not because of errors in splicing and that these variants found in sporadic #Alzheimers were similar to those found in Familial #Alzheimers. CC @alzassociation
It's not just that they realized that the DNA was not germline but genomic cDNA (gencDNA) & was most likely generated by reverse transcriptase, the same enzyme that lets #HIV take over human cells leading to #AIDS@ALZHEIMERSread