Neural networks for genetic epidemiology: past, present, and future
2008

Neural Networks in Genetic Epidemiology

publication Evidence: moderate

Author Information

Author(s): Motsinger-Reif Alison A, Ritchie Marylyn D

Primary Institution: Bioinformatics Research Center, Department of Statistics, North Carolina State University

Hypothesis

Neural networks can improve the identification of disease susceptibility genes in complex diseases.

Conclusion

Neural networks show promise in genetic epidemiology but require further research to optimize their application.

Supporting Evidence

  • Neural networks can handle large amounts of genetic data.
  • They are universal function approximators, meaning they can model complex relationships.
  • Neural networks do not require assumptions about genetic models.

Takeaway

Neural networks are like smart computers that help scientists find genes that make people sick, but they need to be used carefully.

Methodology

The review discusses the application of neural networks in linkage and association analyses for identifying disease susceptibility genes.

Potential Biases

Neural networks may be viewed as 'black boxes', making it difficult to interpret results and understand the influence of input variables.

Limitations

The effectiveness of neural networks in genetic epidemiology is still being explored, and they can produce false positives.

Digital Object Identifier (DOI)

10.1186/1756-0381-1-3

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication