Deciphering causal proteins in Alzheimer's disease: A novel Mendelian randomization method integrated with AlphaFold3 for 3D structure prediction

Minhao Yao,Gary W. Miller,Badri N. Vardarajan,Andrea A. Baccarelli,Zijian Guo,Zhonghua Liu

DOI: https://doi.org/10.1101/2023.02.20.23286200

2024-10-23

Abstract:Hidden confounding biases hinder identifying causal protein biomarkers for Alzheimer's disease in non-randomized studies. While Mendelian randomization (MR) can mitigate these biases using protein quantitative trait loci (pQTLs) as instrumental variables, some pQTLs violate core assumptions, leading to biased conclusions. To address this, we propose MR-SPI, a novel MR method that selects valid pQTL instruments using the Anna Karenina Principle and performs robust post-selection inference. Integrating MR-SPI with AlphaFold3, we developed a computational pipeline to identify causal protein biomarkers and predict 3D structural changes. Applied to genome-wide proteomics data from 54,306 UK Biobank participants and 455,258 subjects (71,880 cases and 383,378 controls) for a genome-wide association study of Alzheimer's disease, we identified seven proteins (TREM2, PILRB, PILRA, EPHA1, CD33, RET, and CD55) with structural alterations due to missense mutations. These findings offer insights into the etiology and potential drug targets for Alzheimer's disease.

Genetic and Genomic Medicine

What problem does this paper attempt to address?

### Problems the Paper Aims to Solve

This paper aims to address the issue of identifying causal protein biomarkers in Alzheimer's disease (AD). Specifically, the paper proposes a new method called MR-SPI, which combines Mendelian Randomization (MR) and AlphaFold3 to overcome potential biases and assumption limitations in existing methods.

### Background and Challenges

1. **Hidden Confounding Bias**:
- In non-randomized studies, hidden confounding factors hinder the identification of causal protein biomarkers for Alzheimer's disease.
- Mendelian Randomization (MR) mitigates these biases by using protein quantitative trait loci (pQTLs) as instrumental variables, but some pQTLs may violate core assumptions, leading to biased conclusions.

2. **Limitations of Existing Methods**:
- Existing MR methods face new challenges when dealing with pQTLs, especially when the number of pQTLs in proteomics data is limited.
- Traditional methods for selecting instrumental variables may not be applicable to different health outcomes, as the underlying genetic architecture may vary by outcome.

### Solution

1. **MR-SPI Method**:
- **Automatic Selection of Valid pQTL Instrumental Variables**: MR-SPI automatically selects valid pQTL instrumental variables through a voting procedure, ensuring these variables meet core assumptions.
- **Robust Post-Selection Inference**: MR-SPI performs robust causal effect estimation after selecting valid instrumental variables and constructs confidence intervals with guaranteed nominal coverage.

2. **3D Structure Prediction**:
- Combining AlphaFold3, MR-SPI predicts 3D structural changes in proteins caused by missense mutations, providing molecular-level biological insights.

### Application and Findings

- **Application Data**: The paper applies whole-genome proteomics data from 54,306 UK Biobank participants and genome-wide association study (GWAS) data from 455,258 subjects (including 71,880 cases and 383,378 controls).
- **Findings**: Using the MR-SPI method, the paper identifies 7 proteins (TREM2, PILRB, PILRA, EPHA1, CD33, RET, and CD55) whose structures are altered due to missense mutations. These findings provide new insights into the etiology of Alzheimer's disease and potential drug targets.

### Conclusion

The paper proposes a novel MR-SPI method that combines Mendelian Randomization and AlphaFold3, successfully identifying causal protein biomarkers for Alzheimer's disease and predicting 3D structural changes in these proteins. These findings not only help in understanding the pathogenesis of Alzheimer's disease but also may provide new directions for developing effective therapeutic interventions.