Predicting the phenotypic effects of non-synonymous single nucleotide polymorphisms based on support vector machines
2007

Predicting the Effects of Genetic Variations

Sample size: 21185 publication Evidence: high

Author Information

Author(s): Tian Jian, Wu Ningfeng, Guo Xuexia, Guo Jun, Zhang Juhua, Fan Yunliu

Primary Institution: Biotechnology Research Institute, Chinese Academy of Agricultural Sciences

Hypothesis

Can support vector machines effectively predict the phenotypic effects of non-synonymous single nucleotide polymorphisms (nsSNPs)?

Conclusion

The performance of Parepro indicates that it is a powerful tool for predicting the effect of nsSNPs on protein function.

Supporting Evidence

  • Parepro achieved a Matthews correlation coefficient (MCC) of 50% and an overall accuracy (Q2) of 76% on the HumVar dataset.
  • The method outperformed existing tools like PolyPhen and SIFT in predicting nsSNP effects.
  • Parepro was able to predict all mutations in the HumVar dataset.

Takeaway

Scientists created a computer program called Parepro that helps figure out how tiny changes in our genes can affect our health.

Methodology

The study used support vector machines to analyze two datasets, HumVar and NewHumVar, to predict the effects of nsSNPs.

Limitations

The method does not incorporate structural information, which may limit its predictive power.

Digital Object Identifier (DOI)

10.1186/1471-2105-8-450

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication