a new method to evaluate the pathogenicity of missense variants
Niigata, Japan – The Department of Neurology at the University of Niigata has developed a new in silico method to evaluate the pathogenicity of missense variants using AlphaFold2 (MOVA). Rare variants in the ALS-causing gene are present in 10–30% of sporadic ALS cases, highlighting the need for accurate and efficient pathogenicity prediction methods. To predict the pathogenicity of a variant, in silico analysis methods are usually used. In some ALS-causing genes, mutations are concentrated in certain regions, and the accuracy of pathogenicity prediction can be improved by considering the positional information of the variants. However, the existing methods do not yet properly take into account the information about the position of the variants in the protein structure. MOVA was developed to address this problem and focuses on using positional information in 3D structures to evaluate the pathogenicity of missense variants. The use of the machine learning method, random forest, in MOVA development has also shown promising results. “Comparison of MOVA with existing in silico analysis methods, such as PolyPhen-2, CADD, REVEL, EVE, and AlphScore, shows its potential in predicting pathogenicity. Combining MOVA with existing methods, such as REVEL and CADD, further enhances performance beyond pathogenicity prediction methods alone. In addition, MOVA also demonstrated superior pathogenicity discrimination of hotspot mutations in the TARDBP and FUS genes. This highlights the importance of considering variant position information in protein structure for better prediction of pathogenicity.” explained Dr. Hatano and Dr. Ishihara. The results of this study were published in the online edition of the journal BMC Bioinformatics in 2023.
Credits: Images can only be used with appropriate captions or credits.
Niigata, Japan – The Department of Neurology at the University of Niigata has developed a new in silico method to evaluate the pathogenicity of missense variants using AlphaFold2 (MOVA). Rare variants in the ALS-causing gene are present in 10–30% of sporadic ALS cases, highlighting the need for accurate and efficient pathogenicity prediction methods. To predict the pathogenicity of a variant, in silico analysis methods are usually used. In some ALS-causing genes, mutations are concentrated in certain regions, and the accuracy of pathogenicity prediction can be improved by considering the positional information of the variants. However, the existing methods do not yet properly take into account the information about the position of the variants in the protein structure. MOVA was developed to address this problem and focuses on using positional information in 3D structures to evaluate the pathogenicity of missense variants. The use of the machine learning method, random forest, in MOVA development has also shown promising results. “Comparison of MOVA with existing in silico analysis methods, such as PolyPhen-2, CADD, REVEL, EVE, and AlphScore, shows its potential in predicting pathogenicity. Combining MOVA with existing methods, such as REVEL and CADD, further enhances performance beyond pathogenicity prediction methods alone. In addition, MOVA also demonstrated superior pathogenicity discrimination of hotspot mutations in the TARDBP and FUS genes. This highlights the importance of considering variant position information in protein structure for better prediction of pathogenicity.” explained Dr. Hatano and Dr. Ishihara. The results of this study were published in the online edition of the journal BMC Bioinformatics in 2023.
Journal
BMC Bioinformatics
DOI
10.1186/s12859-023-05338-5
Article title
Accuracy of machine learning methods based on structural and locational information from AlphaFold2 to predict the pathogenicity of TARDBP and FUS gene variants in ALS
Article Publication Date
19-May-2023
