# The healthy ageing gene expression signature for Alzheimer’s disease diagnosis: a random sampling perspective

## TL;DR

This paper challenges a proposed gene expression signature for Alzheimer's diagnosis by showing it performs no better than random samples.

## Contribution

The novelty is demonstrating that the specific gene set for Alzheimer's diagnosis is not uniquely effective compared to random sets.

## Key findings

- The 150-probe set for Alzheimer's diagnosis does not outperform randomly sampled sets.
- Randomly selected gene sets achieve similar diagnostic performance.
- This suggests the original gene signature lacks special predictive power.

## Abstract

In a recent publication, Sood et al. [2015] presented a set of 150 probe-sets which could be used in a diagnosis of Alzheimer disease (AD) based on gene expression. We reproduce some of their experiments, and show that the performance of their particular set of 150 probe-sets does not stand out compared to that of randomly sampled sets of 150 probe-sets from the same array.