Holiday 2023 Genomics Puzzle
24 views
Skip to first unread message
Don Gilbert
Dec 25, 2023, 2:12:45 PM12/25/23
to EvidentialGene
Solve this puzzle .. and/or read how I did. Your comments are desired.
Cytometric methods of genome size estimation are well established, and are not known to be biased. A majority of recent animal and plant genome assemblies (2020-2023 long-reads) are well below cytometric sizes, from 15% to over 50% below.
Q: What are causes, and cures, of this discrepancy?
A1. cytometry is high-biased, because of .. ?
A2. long-read assembly is low-biased, because of .. ?
A3. genomics is complicated, both measures are right and wrong..
My answer: This trend can be examined through measure of a third evidence, the major contents of the raw DNA sequence used to build these assemblies... read on at
"Measure of major contents in animal and plant genomes, using Gnodes, finds under-assemblies of model plant, Daphnia, fire ant and others."
PS. Biorxiv now has AI-generated summaries from ScienceCast. Do read this if you haven't seen AI doing science writing. After 6 months of my hard work on this paper, following 2 years of genomics data and compute wrangling, it reads like I should have let an Artificial Idiot write this instead. :(
Cytometric methods of genome size estimation are well established, and are not known to be biased. A majority of recent animal and plant genome assemblies (2020-2023 long-reads) are well below cytometric sizes, from 15% to over 50% below.
Q: What are causes, and cures, of this discrepancy?
A1. cytometry is high-biased, because of .. ?
A2. long-read assembly is low-biased, because of .. ?
A3. genomics is complicated, both measures are right and wrong..
My answer: This trend can be examined through measure of a third evidence, the major contents of the raw DNA sequence used to build these assemblies... read on at
"Measure of major contents in animal and plant genomes, using Gnodes, finds under-assemblies of model plant, Daphnia, fire ant and others."
PS. Biorxiv now has AI-generated summaries from ScienceCast. Do read this if you haven't seen AI doing science writing. After 6 months of my hard work on this paper, following 2 years of genomics data and compute wrangling, it reads like I should have let an Artificial Idiot write this instead. :(
Reply all
Reply to author
Forward
0 new messages