Prediction methods have become an integral part of biomedical and biotechnological research. However, their clinical interpretations are largely based on biochemical or molecular data, but not clinical data. Here, we focus on improving the reliability and clinical applicability of prediction algorithms. We assembled and curated two large non-overlapping large databases of clinical phenotypes. These phenotypes were caused by missense variations in 44 and 63 genes associated with Mendelian diseases. We used these databases to establish and validate the model, allowing us to improve the predictions obtained from EVmutation, SNAP2 and PoPMuSiC 2.1. The predictions of clinical effects suffered from a lack of specificity, which appears to be the common constraint of all recently used prediction methods, although predictions mediated by these methods are associated with nearly absolute sensitivity. We introduced evidence-based tailoring of the default settings of the prediction methods; this tailoring substantially improved the prediction outcomes. Additionally, the comparisons of the clinically observed and theoretical variations led to the identification of large previously unreported pools of variations that were under negative selection during molecular evolution. The evolutionary variation analysis approach described here is the first to enable the highly specific identification of likely disease-causing missense variations that have not yet been associated with any clinical phenotype.

Charles University 3rd Faculty of Medicine Prague Czech Republic

See more in PubMed

Biesecker LG, Green RC. Diagnostic clinical genome and exome sequencing. N. Engl. J. Med. 2014;371:1170. doi: 10.1056/NEJMc1409040. PubMed DOI

Simm F, et al. Identification of SLC20A1 and SLC15A4 among other genes as potential risk factors for combined pituitary hormone deficiency. Genet. Med. 2018;20:728–736. doi: 10.1038/gim.2017.165. PubMed DOI

Tennessen JA, et al. Evolution and functional impact of rare coding variation from deep sequencing of human exomes. Science. 2012;337:64–69. doi: 10.1126/science.1219240. PubMed DOI PMC

The 1000 Genomes Project Consortium An integrated map of genetic variation from 1,092 human genomes. Nature. 2012;491:56–65. doi: 10.1038/nature11632. PubMed DOI PMC

Ioannidis NM, et al. REVEL: an ensemble method for predicting the pathogenicity of rare missense variants. Am. J. Hum. Genet. 2016;99:877–885. doi: 10.1016/j.ajhg.2016.08.016. PubMed DOI PMC

Cirulli ET, Goldstein DB. Uncovering the roles of rare variants in common disease through whole-genome sequencing. Nat. Rev. Genet. 2010;11:415–425. doi: 10.1038/nrg2779. PubMed DOI

Bamshad MJ, et al. Exome sequencing as a tool for Mendelian disease gene discovery. Nat. Rev. Genet. 2011;12:745–755. doi: 10.1038/nrg3031. PubMed DOI

Šimčíková D, Kocková L, Vackářová K, Těšínský M, Heneberg P. Evidence-based tailoring of bioinformatics approaches to optimize methods that predict the effects of nonsynonymous amino acid substitutions in glucokinase. Sci. Rep. 2017;7:9499. doi: 10.1038/s41598-017-09810-0. PubMed DOI PMC

Hayat S, Sander C, Marks DS, Elofsson A. All-atom 3D structure prediction of transmembrane β-barrel proteins from sequences. Proc. Natl. Acad. Sci. USA. 2015;110:5413–5418. doi: 10.1073/pnas.1419956112. PubMed DOI PMC

Wang Y, Barth P. Evolutionary-guided de novo structure prediction of self-associated transmembrane helical proteins with near-atomic accuracy. Nat. Commun. 2015;6:7196. doi: 10.1038/ncomms8196. PubMed DOI PMC

Peled S, et al. De-novo protein function prediction using DNA binding and RNA binding proteins as a test case. Nat. Commun. 2016;7:13424. doi: 10.1038/ncomms13424. PubMed DOI PMC

Huang YF, Gulko B, Siepel A. Fast, scalable prediction of deleterious noncoding variants from functional and population genomic data. Nat. Genet. 2017;49:618–624. doi: 10.1038/ng.3810. PubMed DOI PMC

Dehouck Y, Kwasigroch JM, Gilis D, Rooman M. PoPMuSiC 2.1: a web server for the estimation of protein stability changes upon mutation and sequence optimality. BMC Bioinform. 2011;12:151. doi: 10.1186/1471-2105-12-151. PubMed DOI PMC

Hopf TA, et al. Mutation effects predicted from sequence co-variation. Nat. Biotechnol. 2017;35:128–135. doi: 10.1038/nbt.3769. PubMed DOI PMC

Bromberg Y, Kahn PC, Rost B. Neutral and weakly nonneutral sequence variants may define individuality. Proc. Natl. Acad. Sci. USA. 2013;110:14255–14260. doi: 10.1073/pnas.1216613110. PubMed DOI PMC

Vaser R, Adusumalli S, Leng SN, Sikic M, Ng PC. SIFT missense predictions for genomes. Nat. Protoc. 2016;11:1–9. doi: 10.1038/nprot.2015.123. PubMed DOI

Libbrecht MW. Machine learning in genetics and genomics. Nat. Rev. Genet. 2015;16:321–332. doi: 10.1038/nrg3920. PubMed DOI PMC

Sela I, Ashkenazy H, Katoh K, Pupko T. GUIDANCE2: accurate detection of unreliable alignment regions accounting for the uncertainty of multiple parameters. Nucl. Acids Res. 2015;43:W7–W14. doi: 10.1093/nar/gkv318. PubMed DOI PMC

Adebali O, Reznik AO, Ory DS, Zhulin IB. Establishing the precise evolutionary history of a gene improves prediction of disease-causing missense mutations. Genet. Med. 2016;18:1029–1036. doi: 10.1038/gim.2015.208. PubMed DOI PMC

Hecht M, Bromberg Y, Rost B. Better prediction of functional effects for sequence variants. BMC Genom. 2015;16(Suppl 8):S1. doi: 10.1186/1471-2164-16-S8-S1. PubMed DOI PMC

DePristo MA, Weinreich DM, Hartl DL. Missense meanderings in sequence space: a biophysical view of protein evolution. Nat. Rev. Genet. 2005;6:678–687. doi: 10.1038/nrg1672. PubMed DOI

de Visser JA, Krug J. Empirical fitness landscapes and the predictability of evolution. Nat. Rev. Genet. 2014;15:480–490. doi: 10.1038/nrg3744. PubMed DOI

Breen MS, Kemena C, Vlasov PK, Notredame C, Kondrashov FA. Epistasis as the primary factor in molecular evolution. Nature. 2012;490:535–538. doi: 10.1038/nature11510. PubMed DOI

Nykamp K, et al. Sherloc: a comprehensive refinement of the ACMG-AMP variant classification criteria. Genet. Med. 2017;19:1105–1117. doi: 10.1038/gim.2017.37. PubMed DOI PMC

Richards S, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet. Med. 2015;17:405–424. doi: 10.1038/gim.2015.30. PubMed DOI PMC

Stenson PD, et al. The Human Gene Mutation Database: building a comprehensive mutation repository for clinical and molecular genetics, diagnostic testing and personalized genomic medicine. Hum. Genet. 2014;133:1–9. doi: 10.1007/s00439-013-1358-4. PubMed DOI PMC

Liu L, et al. High-density SNP genotyping to define beta-globin locus haplotypes. Blood Cells Mol. Dis. 2009;42:16–24. doi: 10.1016/j.bcmd.2008.07.002. PubMed DOI PMC

Steele AM, et al. The previously reported T342P GCK missense variant is not a pathogenic mutation causing MODY. Diabetologia. 2011;54:2202–2205. doi: 10.1007/s00125-011-2194-5. PubMed DOI

Chellapa K, et al. Src tyrosine kinase phosphorylation of nuclear receptor HNF4α correlates with isoform-specific loss of HNF4α in human colon cancer. Proc. Natl. Acad. Sci. USA. 2012;109:2302–2307. doi: 10.1073/pnas.1106799109. PubMed DOI PMC

Houlleberghs H, et al. Oligonucleotide-directed mutagenesis screen to identify pathogenic Lynch syndrome-associated MSH2 DNA mismatch repair gene variants. Proc. Natl. Acad. Sci. USA. 2016;113:4128–4133. doi: 10.1073/pnas.1520813113. PubMed DOI PMC

Maxwell KN, et al. Evaluation of ACMG-guideline based variant classification of cancer susceptibility and non-cancer-associated genes in families affected by breast cancer. Am. J. Hum. Genet. 2016;98:801–817. doi: 10.1016/j.ajhg.2016.02.024. PubMed DOI PMC

Walsh R, et al. Reassessment of Mendelian gene pathogenicity using 7,855 cardiomyopathy cases and 60,706 reference samples. Genet. Med. 2016;19:192–203. doi: 10.1038/gim.2016.90. PubMed DOI PMC

Hicks S, Wheeler DA, Plon SE, Kimmel M. Prediction of missense mutation functionality depends on both the algorithm and sequence alignment employed. Hum. Mutat. 2011;32:661–668. doi: 10.1002/humu.21490. PubMed DOI PMC

Riera C, Padilla N, de la Cruz X. The complementarity between protein-specific and general pathogenicity predictors for amino acid substitutions. Hum. Mutat. 2016;37:1013–1024. doi: 10.1002/humu.23048. PubMed DOI

Mathe E, et al. Computational approaches for predicting the biological effect of p53 missense mutations: a comparison of three sequence analysis based methods. Nucl. Acids Res. 2006;34:1317–1325. doi: 10.1093/nar/gkj518. PubMed DOI PMC

Pawson T. Protein modules and signaling networks. Nature. 1995;373:573–580. doi: 10.1038/373573a0. PubMed DOI

Aronson HE, Royer WE, Hendrickson WA. Quantification of tertiary structural conservation despite primary sequence drift in the globin fold. Protein Sci. 1994;3:1706–1711. doi: 10.1002/pro.5560031009. PubMed DOI PMC

Rost B. Enzyme function less conserved than anticipated. J. Mol. Biol. 2002;318:595–608. doi: 10.1016/S0022-2836(02)00016-5. PubMed DOI

Miller ML, et al. Pan-cancer analysis of mutation hotspots in protein domains. Cell Systems. 2015;1:197–209. doi: 10.1016/j.cels.2015.08.014. PubMed DOI PMC

Salgado D, et al. UMD-Predictor: a high-throughput sequencing compliant system for pathogenicity prediction of any human cDNA substitution. Hum. Mutat. 2016;37:439–446. doi: 10.1002/humu.22965. PubMed DOI PMC

Havrilla JM, Pedersen BS, Layer RM, Quinlan AR. A map of constrained coding regions in the human genome. Nat. Genet. 2019;51:88–95. doi: 10.1038/s41588-018-0294-6. PubMed DOI PMC

Bastarache L, et al. Phenotype risk scores identify patients with unrecognized Mendelian disease patterns. Science. 2018;359:1233–1239. doi: 10.1126/science.aal4043. PubMed DOI PMC

Romeo S, et al. Rare loss-of-function mutations in ANGPTL family members contribute to plasma triglyceride levels in humans. J. Clin. Invest. 2009;119:70–79. PubMed PMC

Rost B, Radivojac P, Bromberg Y. Protein function in precision medicine: deep understanding with machine learning. FEBS Lett. 2016;590:2327–2341. doi: 10.1002/1873-3468.12307. PubMed DOI PMC

Fowler DM, Fields S. Deep mutational scanning: a new style of protein science. Nat. Methods. 2014;11:801–807. doi: 10.1038/nmeth.3027. PubMed DOI PMC

Boucher JI, Bolon DN, Tawfik DS. Quantifying and understanding the fitness effects of protein mutations: Laboratory versus nature. Protein Sci. 2016;25:1219–1226. doi: 10.1002/pro.2928. PubMed DOI PMC

Singleton MV, et al. Phevor combines multiple biomedical ontologies for accurate identification of disease-causing alleles in single individuals and small nuclear families. Am. J. Hum. Genet. 2014;94:599–610. doi: 10.1016/j.ajhg.2014.03.010. PubMed DOI PMC

Bone WP, et al. Computational evaluation of exome sequence data using human and model organism phenotypes improves diagnostic efficiency. Genet. Med. 2016;18:608–617. doi: 10.1038/gim.2015.137. PubMed DOI PMC

Simonti CN, et al. The phenotypic legacy of admixture between modern humans and Neadertals. Science. 2016;351:737–741. doi: 10.1126/science.aad2149. PubMed DOI PMC

Posey JE, et al. Resolution of disease phenotypes resulting from multilocus genomic variation. N. Engl. J. Med. 2017;376:21–31. doi: 10.1056/NEJMoa1516767. PubMed DOI PMC

Bendl J, et al. PredictSNP: robust and accurate consensus classifier for prediction of disease-related mutations. PLoS Comput. Biol. 2014;10:e1003440. doi: 10.1371/journal.pcbi.1003440. PubMed DOI PMC

Bendl J, et al. PredictSNP2: a unified platform for accurately evaluating SNP effects by exploiting the different characteristics of variants in distinct genomic regions. PLoS Comput. Biol. 2016;12:e1004962. doi: 10.1371/journal.pcbi.1004962. PubMed DOI PMC

Kircher M, et al. A general framework for estimating the relative pathogenicity of human genetic variants. Nat. Genet. 2014;46:310–315. doi: 10.1038/ng.2892. PubMed DOI PMC

Quang D, Chen Y, Xie X. DANN: a deep learning approach for annotating the pathogenicity of genetic variants. Bioinformatics. 2015;31:761–763. doi: 10.1093/bioinformatics/btu703. PubMed DOI PMC

Sarkozy A, et al. Clinical and molecular analysis of 30 patients with multiple lentigines LEOPARD syndrome. J. Med. Genet. 2004;41:e68. doi: 10.1136/jmg.2003.013466. PubMed DOI PMC

Yoshida R, et al. Two novel and one recurrent PTPN11 mutations in LEOPARD syndrome. Am. J. Med. Genet. A. 2004;130A:432–434. doi: 10.1002/ajmg.a.30281. PubMed DOI

Osawa R, et al. A novel PTPN11 missense mutation in a patient with LEOPARD syndrome. Br. J. Dermatol. 2009;161:1202–1204. doi: 10.1111/j.1365-2133.2009.09385.x. PubMed DOI

Digilio MC, et al. Grouping of multiple-lentigines/LEOPARD and Noonan syndromes on the PTPN11 gene. Am. J. Hum. Genet. 2002;71:389–394. doi: 10.1086/341528. PubMed DOI PMC

Seishima M, et al. Malignant melanoma in a woman with LEOPARD syndrome: identification of a germline PTPN11 mutation and a somatic BRAF mutation. Br. J. Dermatol. 2007;157:1297–1299. doi: 10.1111/j.1365-2133.2007.08229.x. PubMed DOI

Conti E, et al. A novel PTPN11 mutation in LEOPARD syndrome. Hum. Mutat. 2003;21:654. doi: 10.1002/humu.9149. PubMed DOI

Keren B, et al. PTPN11 mutations in patients with LEOPARD syndrome: a French multicentric experience. J. Med. Genet. 2004;41:e117. doi: 10.1136/jmg.2004.021451. PubMed DOI PMC

Sarkozy A, et al. Correlation between PTPN11 gene mutations and congenital heart defects in Noonan and LEOPARD syndromes. J. Med. Genet. 2003;40:704–708. doi: 10.1136/jmg.40.9.704. PubMed DOI PMC

Atik T, et al. Mutation spectrum and phenotypic features in Noonan syndrome with PTPN11 mutations: definition of two novel mutations. Indian J. Pediatr. 2016;83:517–521. doi: 10.1007/s12098-015-1998-6. PubMed DOI

Tartaglia M, et al. PTPN11 mutations in Noonan syndrome: molecular spectrum, genotype-phenotype correlation, and phenotypic heterogeneity. Am. J. Hum. Genet. 2002;70:1555–1563. doi: 10.1086/340847. PubMed DOI PMC

Al-Gazali L, Ali BR. Mutations of a country: a mutation review of single gene disorders in the United Arab Emirates (UAE) Hum. Mutat. 2010;31:505–520. doi: 10.1002/humu.21232. PubMed DOI

Knott M, et al. Novel and Mediterranean beta thalassemia mutations in the indigenous Northern Ireland population. Blood Cells Mol. Dis. 2006;36:265–268. doi: 10.1016/j.bcmd.2005.12.005. PubMed DOI

Colah R, et al. Regional heterogeneity of beta-thalassemia mutations in the multi ethnic Indian population. Blood Cells Mol. Dis. 2009;42:241–246. doi: 10.1016/j.bcmd.2008.12.006. PubMed DOI

Villegas A, et al. Hb Santander [beta34(B16)Val–> Asp (GTC–> GAC)]: a new unstable variant found as a de novo mutation in a Spanish patient. Hemoglobin. 2003;27:31–35. doi: 10.1081/HEM-120016378. PubMed DOI

Henderson SJ, et al. Ten years of routine α- and β-globin gene sequencing in UK hemoglobinopathy referrals reveals 60 novel mutations. Hemoglobin. 2016;40:75–84. doi: 10.3109/03630269.2015.1113990. PubMed DOI

Zanella-Cleon I, et al. Strategy for identification by mass spectrometry of a new human hemoglobin variant with two mutations in Cis in the beta-globin chain: Hb S-Clichy [beta6(A3)Glu–>Val; beta8(A5)Lys–>Thr] Hemoglobin. 2009;33:177–187. doi: 10.1080/03630260903061184. PubMed DOI

Wajcman H, et al. Two new hemoglobin variants with increased oxygen affinity: Hb Nantes [beta34(B16)Val–>Leu] and Hb Vexin [beta116(G18)His–>Leu] Hemoglobin. 2003;27:191–199. doi: 10.1081/HEM-120023384. PubMed DOI

McClure RF, Hoyer JD, Mai M. The JAK2 V617F mutation is absent in patients with erythrocytosis due to high oxygen affinity hemoglobin variants. Hemoglobin. 2006;30:487–489. doi: 10.1080/03630260600868147. PubMed DOI

Shin SY, Bang SM, Kim HJ. A novel hemoglobin variant associated with congenital erythrocytosis: Hb Seoul [β86(F2)Ala→Thr] (HBB:c.259G>A) Ann. Clin. Lab. Sci. 2016;46:312–314. PubMed

Vulliamy T, Beutler E, Luzzatto L. Variants of glucose-6-phosphate dehydrogenase are due to missense mutations spread throughout the coding region of the gene. Hum. Mutat. 1993;2:159–167. doi: 10.1002/humu.1380020302. PubMed DOI

Bulliamy T, Luzzatto L, Hirono A, Beutler E. Hematologically important mutations: glucose-6-phosphate dehydrogenase. Blood Cells Mol. Dis. 1997;23:302–313. doi: 10.1006/bcmd.1997.0147. PubMed DOI

Yan T, et al. Incidence and complete molecular characterization of glucose-6-phosphate dehydrogenase deficiency in the Guangxi Zhuang autonomous region of southern China: description of four novel mutations. Haematologica. 2006;91:1321–1328. PubMed

McGlacken-Byrne SM, et al. The evolving course of HNF4A hyperinsulinaemic hypoglycaemia–a case series. Diabet. Med. 2014;31:e1–e5. doi: 10.1111/dme.12259. PubMed DOI

Flanagan SE, et al. Diazoxide-responsive hyperinsulinemic hypoglycemia caused by HNF4A gene mutations. Eur. J. Endocrinol. 2010;162:987–992. doi: 10.1530/EJE-09-0861. PubMed DOI PMC

Colclough K, Bellanne-Chantelot C, Saint-Martin C, Flanagan SE, Ellard S. Mutations in the genes encoding the transcription factors hepatocyte nuclear factor 1 alpha and 4 alpha in maturity-onset diabetes of the young and hyperinsulinemic hypoglycemia. Hum. Mutat. 2013;34:669–685. doi: 10.1002/humu.22279. PubMed DOI

Urbanová J, et al. Positivity for islet cell autoantibodies in patients with monogenic diabetes is associated with later diabetes onset and higher HbA1c level. Diabet. Med. 2014;31:466–471. doi: 10.1111/dme.12314. PubMed DOI

Harries LW, et al. The diabetic phenotype in HNF4A mutation carriers is moderated by the expression of HNF4A isoforms from the P1 promoter during fetal development. Diabetes. 2008;57:1745–1752. doi: 10.2337/db07-1742. PubMed DOI

Song S, et al. EDA gene mutations underlie non-syndromic oligodontia. J. Dent. Res. 2009;88:126–131. doi: 10.1177/0022034508328627. PubMed DOI PMC

Lee KE, et al. Oligodontia and curly hair occur with ectodysplasin-a mutations. J. Dent. Res. 2014;93:371–375. doi: 10.1177/0022034514522059. PubMed DOI

Ruiz-Heiland G, et al. Novel missense mutation in the EDA gene in a family affected by oligodontia. J. Orofac. Orthop. 2016;77:31–38. doi: 10.1007/s00056-015-0005-1. PubMed DOI

Cluzeau C, et al. Only four genes (EDA1, EDAR, EDARADD, and WNT10A) account for 90% of hypohidrotic/anhidrotic ectodermal dysplasia cases. Hum. Mutat. 2011;32:70–72. doi: 10.1002/humu.21384. PubMed DOI

Guazzarotti L, et al. Phenotypic heterogeneity and mutational spectrum in a cohort of 45 Italian males subjects with X-linked ectodermal dysplasia. Clin. Genet. 2015;87:338–342. doi: 10.1111/cge.12404. PubMed DOI

Clauss F, et al. X-linked and autosomal recessive Hypohidrotic Ectodermal Dysplasia: genotypic-dental phenotypic findings. Clin. Genet. 2010;78:257–266. doi: 10.1111/j.1399-0004.2010.01376.x. PubMed DOI

Monreal AW, Zonana J, Ferguson B. Identification of a new splice form of the EDA1 gene permits detection of nearly all X-linked hypohidrotic ectodermal dysplasia mutations. Am. J. Hum. Genet. 1998;63:380–389. doi: 10.1086/301984. PubMed DOI PMC

Schneider P, et al. Mutations leading to X-linked hypohidrotic ectodermal dysplasia affect three major functional domains in the tumor necrosis factor family member ectodysplasin-A. J. Biol. Chem. 2001;276:18819–18827. doi: 10.1074/jbc.M101280200. PubMed DOI

Pääkkönen K, et al. The mutation spectrum of the EDA gene in X-linked anhidrotic ectodermal dysplasia. Hum. Mutat. 2001;17:349. doi: 10.1002/humu.33. PubMed DOI