Class 4 Discussion Seeds

Stephens 2025

1. Is the categorization of "acquisition, storage, distribution, and analysis" a good way to think about data and data growth? In your own work or your lab's work, which of these four is the actual bottleneck? Or is it something not mentioned at all, like your own time and knowledge of what to do?

2. Crystal ball gazing is difficult! Here we are and we can score the predictions. They got some things right about growth:

• Then: "currently there are more than 2,500 high-throughput instruments, manufactured by several different companies"
  • Now: Illumina alone reports more than 25k instruments
• Then: "SRA currently contains more than 3.6 petabases of raw sequence data"
  • Now: 107 petabases And some things very wrong. Why?
• Then: "We therefore estimate between 100 million and as many as 2 billion human genomes could be sequenced by 2025"
  • Now: "At least around 2 million humans, likely more, have had their whole genomes sequenced." cite.
  • Things to consider: supply vs demand, cost, utility

3. What happened to these ideas? Why didn't they happen (or did they?) Are they still good goals?

• "More significant reductions in storage demand will come when improvements in sequencing accuracy and database comprehensiveness reach the point at which genome sequences themselves do not need to be stored, just the list of variants relative to a reference collection (“delta encoding”)" (Things to consider: quality scores, references, cram)
• "we anticipate that raw RNA-seq reads will no longer be stored, except for specific research purposes"
• "The most practical, and perhaps only, solution for distributing genome sequences at a population scale is to use cloud-computing systems that minimize data movement and maximize code federation"
• "we anticipate the development of huge genomics archives used for storing millions of genomes along with the associated‘omics measurements over time. Ideally, these archives will also collect or be linked to the patient phenotypic data, especially disease outcomes and treatments provided to support retrospective analysis as new relationships are discovered."

BLAST and DIAMOND

1. You BLAST a sequence in 2019 and get $$E = 0.001$$. You rerun the identical search in 2026 against the current database and get $$E = 0.02$$. The alignment is byte-for-byte identical. What changed, and which result should you report?

2. Your protein has a 30-residue insertion relative to its closest homolog. Sketch what BLAST 1990 reports versus BLAST 1997. Which is easier to interpret, and why does it matter for annotation?

3. DIAMOND doesn't use a fundamentally better algorithm than BLAST. Where does the 20,000× actually come from? What does that suggest about how you should think about performance in your own analyses?

4. Two sequences share 95% identity over 12 residues. Two others share 28% identity over 400 residues. Which is more likely to be a genuine homolog, and which statistic tells you?

5. Explain, without using the words "p-value" or "significant," what $$E = 10^{-8}$$ means to a colleague who has never run BLAST.