Replies to post #508963 on Anavex Life Sciences Corp (AVXL)

[boi568]

I had to follow up on my initial question, since it was clear that ChatGTP was still referring to the entire human genome. We are now down to an estimate of only 150 to 350 genes sampled by Anavex. Without taking into account the other correlations discussed in Hosai's post, but just applying the Bonferroni correction (controlling for multiplicity) to the ABCLEAR p value of .0004, we get an estimated p value of between .06 to .14, despite no prespecificity.

I think Anavex can make a strong argument to CHMP in support of the ABCLEAR3 results.

Here's that Q&A:

Q. "No. Anavex was not referring to the prespecified group. It was referring to a subset of the human genome based on the qualifications I gave you. You are not applying qualifications to your answer."

A. Thank you — this is an important clarification, and it changes the interpretation.

You are saying:

Anavex did not screen the entire genome.
They screened only a subset of genes that were:

AD-relevant

Related to the mechanism of blarcamesine (s-1 receptor / mitochondrial stress / proteostasis, etc.)

Pre-specified (but not the tiny SIGMAR1/COMT subgroup)

So your question becomes:

Given those constraints, how many genes would realistically fall into an
“AD-relevant & s-1/mechanism-related” subset?

This is much smaller than the whole genome and much smaller than the 27k–33k gene-locus discovery set.

Let’s estimate that subset.

✅ Step-by-step reasoning
1. AD-relevant gene catalogs

Well-curated AD-related gene databases include roughly:

~650–850 AD-associated genes (AlzGene, GeneCards AD pathway sets, KEGG AD pathway, Reactome AD modules)

This includes all genes tied to:

amyloid processing

tau phosphorylation

neuroinflammation

synaptic function

oxidative stress

mitochondrial pathways

So baseline AD-relevant genomic space ˜ 700 ± 200 genes.

2. Mechanism-related genes for blarcamesine (s-1 receptor)

The s-1 receptor pathway touches:

ER stress / UPR

mitochondrial function

calcium homeostasis

BDNF / neurotrophin pathways

autophagy / proteostasis

cholinergic signaling

oxidative stress response

These pathways collectively involve roughly:

~300–500 genes, depending on annotation set.

3. Intersection of (AD-relevant) n (s-1 / mechanism-related)

This is the critical constraint.

Overlap between AD-relevant genes and the s-1 mechanism pathways is typically about 20–40% depending on classification.

So expected overlapping gene set:

˜ 150–350 genes

This is the best estimate for the number of genes that meet both criteria you specified.

🎯 Final Estimate
**Anavex likely screened on the order of:

➡️ ~150–350 protein-coding genes**

…when limited to “AD-relevant, mechanism-related, pre-specified variants.”

This is:

far less than the whole genome

far less than the 27k–33k discovery dataset

substantially more than the 1–2 genes used for the efficacy subgroup

And importantly —

✔ This estimate applies your stated qualification

(“subset of the genome relevant to AD and the drug’s mechanism”).