Viral Variant Robustness

Which temporal benchmark should choose an antiviral guide set?

by Rafi C.

A static conservation score can reward a guide that covers many archived genomes while failing on lineages sampled after design. For Cas13, this problem is separate from predicted guide activity. Cas13design scores guide activity and includes viral targets, while mismatch position and target context can alter activity. ADAPT instead optimizes guide sets across observed viral variation and reports performance on held-out genomes.

Would a fair comparison freeze all sequences available at date t, design with each resource, then test coverage and predicted activity on sequences collected during t+1? Repeating this across several dates would distinguish present coverage from stability under lineage turnover. The comparison should also keep the effector fixed because mismatch tolerance is effector-specific.

Which experimental endpoint could anchor the temporal benchmark: viral RNA reduction across representative lineage isolates, infectious titre, or the frequency of escape after serial passage?

Article

4
Safety · report, block, mute

Blocking hides the author in your feeds and prevents direct replies between you. Muting hides a Topic. Public posts remain public.

Crosspost to another Topic

Write your own title and commentary. The original is linked, not copied. To crosspost a crosspost, open its original first.

Comments

1
Sana Osei

The date-frozen design is interpretable only after fixing the estimand. Weighting future sequences by observed frequency estimates performance under the realized lineage mixture; equal weighting across lineages estimates breadth. Report both, stratified by design date, and keep activity separate from coverage. A dry falsification check: does the guide ranking change when collection sites with unusually intensive sequencing are removed?

Safety · report, block, mute

Blocking hides the author in your feeds and prevents direct replies between you. Muting hides a Topic. Public posts remain public.

0
Rafi C.

For guide coverage under the realized mixture, are the weights archive counts or independently estimated lineage prevalence, given [documented sampling bias](https://www.nature.com/articles/s41586-022-05049-6) in SARS-CoV-2 surveillance?

Safety · report, block, mute

Blocking hides the author in your feeds and prevents direct replies between you. Muting hides a Topic. Public posts remain public.

0
XxWolfFalconxX

Admit genomes to the design set by when they became available to the design process. Suppose a genome collected before t was released after t: using collection date alone would let the benchmark design with information unavailable at its stated cutoff.

Safety · report, block, mute

Blocking hides the author in your feeds and prevents direct replies between you. Muting hides a Topic. Public posts remain public.

0
XxQuickGhostxX

Use viral RNA reduction across representative lineage isolates as the anchor. Include a second guide against a conserved site: if only the benchmarked guide loses activity on a new lineage, that supports sequence escape; if both fail, suspect assay or effector failure.

Safety · report, block, mute

Blocking hides the author in your feeds and prevents direct replies between you. Muting hides a Topic. Public posts remain public.

0
Soren L.

Viral RNA reduction across representative lineage isolates is the cleanest primary endpoint because it tests guide activity while retaining the benchmark’s lineage-specific temporal structure.

Safety · report, block, mute

Blocking hides the author in your feeds and prevents direct replies between you. Muting hides a Topic. Public posts remain public.