iBet uBet web content aggregator. Adding the entire web to your favor.
iBet uBet web content aggregator. Adding the entire web to your favor.



Link to original content: http://pubmed.ncbi.nlm.nih.gov/38255924/
Classification of MLH1 Missense VUS Using Protein Structure-Based Deep Learning-Ramachandran Plot-Molecular Dynamics Simulations Method - PubMed Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Jan 10;25(2):850.
doi: 10.3390/ijms25020850.

Classification of MLH1 Missense VUS Using Protein Structure-Based Deep Learning-Ramachandran Plot-Molecular Dynamics Simulations Method

Benjamin Tam  1   2   3 , Zixin Qin  1   2   3 , Bojin Zhao  1   2   3 , Siddharth Sinha  1   2   3 , Chon Lok Lei  1   2   3 , San Ming Wang  1   2   3
Affiliations

Classification of MLH1 Missense VUS Using Protein Structure-Based Deep Learning-Ramachandran Plot-Molecular Dynamics Simulations Method

Benjamin Tam et al. Int J Mol Sci. .

Abstract

Pathogenic variation in DNA mismatch repair (MMR) gene MLH1 is associated with Lynch syndrome (LS), an autosomal dominant hereditary cancer. Of the 3798 MLH1 germline variants collected in the ClinVar database, 38.7% (1469) were missense variants, of which 81.6% (1199) were classified as Variants of Uncertain Significance (VUS) due to the lack of functional evidence. Further determination of the impact of VUS on MLH1 function is important for the VUS carriers to take preventive action. We recently developed a protein structure-based method named "Deep Learning-Ramachandran Plot-Molecular Dynamics Simulation (DL-RP-MDS)" to evaluate the deleteriousness of MLH1 missense VUS. The method extracts protein structural information by using the Ramachandran plot-molecular dynamics simulation (RP-MDS) method, then combines the variation data with an unsupervised learning model composed of auto-encoder and neural network classifier to identify the variants causing significant change in protein structure. In this report, we applied the method to classify 447 MLH1 missense VUS. We predicted 126/447 (28.2%) MLH1 missense VUS were deleterious. Our study demonstrates that DL-RP-MDS is able to classify the missense VUS based solely on their impact on protein structure.

Keywords: MLH1; Ramachandran plot; VUS; autoencoder; deep learning; molecular dynamics simulation; neural network.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Distribution of benign and pathogenic variants in the MLH1 N-terminus (residues 1–347). The locations of benign and pathogenic variants were marked. The benign variants were enriched in residues 18–141 and 213–347, whereas the pathogenic variants were enriched in residues 19–155 and 226–346. Blue: variant position.
Figure 2
Figure 2
The violin distribution plot of 1μs wildtype RMSD. The MutS-HI domain caused the most significant change in RMSD. At 0.75 nm, the α-helix I and H in the MutS-HI domain showed a more compact (closed) structure. Whereas the other RMSD distance, the α-helix I and H, dissociated with higher structure flexibility. Red: the α-helix I and H; Tan: structure of MLH1; Teal: distributions of RMSD.
Figure 3
Figure 3
Illustration of an autoencoder and multi-layer classifier. DL-RP-MDS used the wildtype, benign, and pathogenic Ramachandran scatter plot as the input data. The model was optimized with three hidden layers in the autoencoder. The six latent representation dimensions were used in the multi-layer classifier, and the model output the probability of deleterious and unknown for the variants.
Figure 4
Figure 4
Latent dimensions generated by DL-RP-MDS model. Ramachandran scatter plots were reduced by autoencoder and retained dense-information dimensions. The overlapped regions between benign, WT, and pathogenic variants showed the common structural features of the variants, whereas the non-overlapped regions represented the structural features caused by the variants. The criteria for classification were based on the combination of each latent dimension. Blue: Benign; Red: Pathogenic; Green: wildtype allele.
Figure 5
Figure 5
Distribution of deleterious VUS variants by DL-RP-MDS model. The orange lollipop represents the location of predicted deleterious variants in MLH1. Deleterious variants with p(D) > 85% were named in the figure. Orange: deleterious variant position.
Figure 6
Figure 6
Structural change in MLH1 by G181D and V326M. (a) The wildtype G181 interacted with 4 residues (L177, S184, V185, and I219), whereas the variant D181 interacted with 5 residues (L177, E178, S184, V185, and I219). (b) The D181 caused instability of αG and further affected the MutS-HI domain. (c) The wildtype V326 interacted with 9 residues (F240, M242, L272, I322, N329, I330, S340, R341, and M342), whereas the M326 interacted with altered 9 residues (L272, I322, L323, R325, Q328, I330, S340, M342, and Y343). (d) The M326 did not interact with the β sheet and caused the αI helix to detach. Grey: wildtype; peach: variant; green: interacting atoms; purple: non-interacting atoms; red label: wildtype and variants; black label: interacting residues.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Dai J., Sanchez A., Adam C., Ranjha L., Reginato G., Chervy P., Tellier-Lebegue C., Andreani J., Guérois R., Ropars V., et al. Molecular basis of the dual role of the Mlh1-Mlh3 endonuclease in MMR and in meiotic crossover formation. Proc. Natl. Acad. Sci. USA. 2021;118:e2022704118. doi: 10.1073/pnas.2022704118. - DOI - PMC - PubMed
    1. Jia P., Chai W. The MLH1 ATPase domain is needed for suppressing aberrant formation of interstitial telomeric sequences. DNA Repair. 2018;65:20–25. doi: 10.1016/j.dnarep.2018.03.002. - DOI - PMC - PubMed
    1. Ryan N.A.J., Glaire M.A., Blake D., Cabrera-Dandy M., Evans D.G., Crosbie E.J. The proportion of endometrial cancers associated with Lynch syndrome: A systematic review of the literature and meta-analysis. Genet. Med. 2019;21:2167–2180. doi: 10.1038/s41436-019-0536-8. - DOI - PMC - PubMed
    1. Lynch H.T., de la Chapelle A. Hereditary Colorectal Cancer. N. Engl. J. Med. 2003;348:919–932. doi: 10.1056/NEJMra012242. - DOI - PubMed
    1. Dusic E.J., Theoryn T., Wang C., Swisher E.M., Bowen D.J., EDGE Study Team Barriers, interventions, and recommendations: Improving the genetic testing landscape. Front. Digit. Health. 2022;4:961128. doi: 10.3389/fdgth.2022.961128. - DOI - PMC - PubMed

MeSH terms