Phenotype-aware prioritisation of rare Mendelian disease variants DOI Creative Commons
Catherine Kelly, Anita Szabó, Nikolas Pontikos

et al.

Trends in Genetics, Journal Year: 2022, Volume and Issue: 38(12), P. 1271 - 1283

Published: Aug. 4, 2022

Next-generation sequencing technologies have made achieving a molecular diagnosis for rare genetic disorder more and feasible and, in turn, enabled personalised clinical management of the affected patients their families.Identifying one or two variants that are responsible certain disease phenotype from millions identified by can be time-consuming expensive.Numerous phenotype-aware variant prioritisation (VP) software tools now exist to help semi-automate process diseases.Although many published VP limitations, show lack maintenance, become soon unfit usage, several up-to-date demonstrate an impressive capacity prioritising diagnoses when tested on real patient data.Adopting diagnostics settings efficiently assist multidisciplinary teams clinicians scientists reporting disease. A analysis data Mendelian diseases has huge impact families. Numerous been developed automate this process, shorten diagnostic odyssey, but performance statistics limited. Here we identify, assess, compare all up-to-date, freely available, programmatically accessible using whole-exome, retinal dataset 134 individuals with diagnosis. All were able identify around two-thirds as top-ranked candidate, LIRICAL performing best overall. Finally, discuss challenges overcome most cases remaining undiagnosed after current, state-of-the-art practices. With approximately 80% having origin, identifying correct causative single-gene disorders creates greater potential informed through precision medicine recommendation drug trials, rather than only treating evident symptoms. Improvements information at scale parallelisation (next-generation sequencing) greatly increased quantities genomic production lower overall costs, shown recent completion 100,000 Genomes Project UK [1.Smedley D. et al.100,000 pilot rare-disease health care - preliminary report.N. Engl. J. Med. 2021; 385: 1868-1880Crossref PubMed Scopus (192) Google Scholar]. Whole-exome (WES) is still commonly used method, exome (~2% human genome) harbours ~85% currently known disease-causing sequence [2.Caspar S.M. al.Clinical sequencing: raw lifetime value.Clin. Genet. 2018; 93: 508-519Crossref (64) The candidate typical WES experiment often derived 60 000 100 affecting protein-coding regions, which nearly will benign unrelated [3.De La Vega F.M. al.Artificial intelligence enables comprehensive genome interpretation nomination diseases.Genome 13: 153Crossref (25) However, filtering review involve tens, if not few hundreds, usually both expensive done via manual scientists. Around one-third children born do live see fifth birthday [4.Wright C.F. al.Paediatric genomics: diagnosing children.Nat. Rev. 19: 253-268Crossref (298) Scholar], so it vital rapid yet, traumatic wait time lengthy (e.g., median 6 years Project) offers possibility disease-causative efficiently, sometimes within minutes. These discard large likely benign, common strategies based publicly available gnomAD) in-house databases. vast majority prioritise single-nucleotide (SNVs) small insertion/deletions (indels) formatted call format (VCF) files. To determine SNVs/indels, incorporate existing silico pathogenicity prediction restrict patients' VCF files interest range methods. They include function-prediction methods MutationTaster, PolyPhen-2, SIFT), likelihood each missense causing pathogenic changes protein structure function; phylogenetic conservation GERP++, phastCons, phyloP), measure degree given nucleotide site; other methods, concern tailored use deep neural networks MVP, PrimateAI); ensemble CADD, DANN, REVEL), integrate multiple component [5.Li al.Performance evaluation pathogenicity-computation variants.Nucleic Acids Res. 46: 7793-7804Crossref (139) Despite availability wide tools, improvements needed discriminate reported specificity 65%; furthermore, sensitivities ranging 51% 96% (median, 88%), relying algorithm-predicted generate number false positive candidates aim automating where relevance gene patient's taken into account, virtually incorporation standardised phenotypic terms, drawing 15 terms Human Phenotype Ontology (HPO) [6.Köhler S. al.The 2021.Nucleic 2020; 49: D1207-D1217Crossref (363) This ultimately significant addition; example, Exomiser (among first its kind) [7.Bone W.P. al.Computational model organism phenotypes improves efficiency.Genet. 2016; 18: 608-617Abstract Full Text PDF (63) Scholar,8.Robinson P.N. al.Improved prioritization genes cross-species comparison.Genome 2014; 24: 340-348Crossref (235) Scholar] demonstrated top diagnosed 20–77% (using variant-based filtering) 96–97% (with addition HPO terms) simulated across different mode inheritances (MOIs) well 3% 74% inferred MOIs [9.Cipriani V. al.An improved phenotype-driven tool prioritization: benchmarking exomiser whole-exome data.Genes. 11: 460Crossref (28) date (different) and/or very datasets, limited comparison. Strikingly, specific always claims outperform relatively literature. Here, set out perform thorough literature tools. Building previous then conducted unbiased comparison selected whole-exomes inherited (IRDs) detailed search was carried list would meet following criteria: (i) directly accepting files; (ii) describe phenotypes; (iii) being (last updated since 2018); (iv) academic use; (v) local, programmatic access (and therefore safer opposed web-based allowing processing scale). Literature searches combination keywords (i.e., 'exome', 'genome', 'variant prioritisation', alternative spelling prioritization', 'human ontology') returned about 400 peer-reviewed journal articles (11 March 2022) (Figure 1). Articles screened those publications involved initially gave 37 Scholar,10.Alemán A. al.A interactive framework studies.Nucleic 42: W88-W93Crossref (35) Scholar, 11.Anderson al.Personalised analytics diagnostics.Nat. Commun. 2019; 10: 5274Crossref (10) 12.Antanaviciute al.OVA: integrating physical biomedical domain ontologies enhanced prioritization.Bioinformatics. 2015; 31: 3822-3829PubMed 13.Bertoldi L. al.QueryOR: web platform prioritization.BMC Bioinform. 2017; 225Crossref (18) 14.Birgmeier al.AMELIE speeds matching genotype primary literature.Sci. Transl. 12eaau9113Crossref (30) 15.Bosio M. al.eDiVA-classification diagnostics.Hum. Mutat. 40: 865-878Crossref (12) 16.Boudellioua I. al.DeepPVP: phenotype-based learning.BMC 20: 65Crossref (36) 17.Boudellioua al.Semantic novel variants.PLoS Comput. Biol. 13e1005500Crossref (24) 18.Chiara al.VINYL: Variant prIoritizatioN survivaL analysis.Bioinformatics. 36: 5590-5599Crossref (4) 19.Desvignes J.-P. al.VarAFT: annotation filtration system next generation data.Nucleic W545-W553Crossref (114) 20.Holt J.M. al.VarSight: prioritizing clinically binary classification algorithms.BMC 1-10Crossref (11) 21.Holtgrewe al.VarFish: DNA research.Nucleic 48: W162Crossref 22.Hombach al.MutationDistiller: user-driven identification 47: W114-W120Crossref 23.Hunt S.E. al.Annotating Ensembl Effect Predictor-a tutorial.Hum. 43: 986-997Crossref 24.Ip E. al.VPOT: customizable ordering annotated variants.Genom. Proteom. 17: 540Crossref (7) 25.James R.A. visual curatorial approach discovery genome-wide diagnostics.Genome 8: 13Crossref 26.Javed al.Phen-Gen: combining analyze disorders.Nat. Methods. 935-937Crossref (106) 27.Kennedy B. al.Using VAAST disease-associated next-generation data.Curr. Protoc. Hum. 81: 6PubMed 28.Koile al.GenIO: phenotype-genotype server genomics diseases.BMC 25Crossref (9) 29.Li M.J. al.wKGGSeq: strategy-based disease-targeted online facilitate studies disorders.Hum. 496-503Crossref (8) 30.Li Q. al.Xrare: machine learning method jointly modeling evidence diagnosis.Genet. 21: 2126-2134Abstract (37) 31.Li Z. al.PhenoPro: toolkit assisting disease.Bioinformatics. 35: 3559-3566Crossref (21) 32.Manshaei R. al.GeneTerpret: multilayer interpretation.BMC 15: 31Google 33.Muller H. al.VCF.Filter: disease-linked 45: W567-W572Crossref (19) 34.O'Brien T.D. (AI)-assisted reanalysis aids new reduces laboratory.Genet. 2022; 192-200Abstract 35.Robinson al.Interpretable ratio paradigm.Am. 107: 403-417Abstract 36.Seo G.H. al.Diagnostic yield utility whole automated system, EVIDENCE.Clin. 98: 562-570Crossref (51) 37.Sifrim al.eXtasy: fusion.Nat. 2013; 1083-1084Crossref (131) 38.Singleton M.V. al.Phevor combines accurate alleles single nuclear families.Am. 94: 599-610Abstract (137) 39.Stelzer G. al.VarElect: variation prioritizer GeneCards Suite.BMC Genomics. 444Crossref (126) 40.Trakadis Y.J. al.PhenoVar: polymalformative syndromes.BMC 7: 22Google 41.Ward al.Clin.iobio: collaborative workflow enable team-based genomics.J. Pers. 12: 73Crossref (1) 42.Wu C. al.Rapid exomes Phenoxome: computational approach.Eur. 27: 612-620Crossref (13) 43.Yang Wang K. Genomic ANNOVAR wANNOVAR.Nat. 1556-1566Crossref (566) 44.Zemojtel T. al.Effective genome.Sci. 6252ra123Crossref (188) prune according aforementioned criteria. Remarkably, seven passed five criteria final testing Table 1 shows details retrieved corresponding selection process.Table 1Selection suitability criteriaaTable criteriaaaA grey cell indicates feature present.bFollowing review, viable met terms; last available; access) aA present. bFollowing Most (33) accept file, standard file storing data. total 28 'phenotype-aware' simply prioritisation; they allow integrative HPO, de facto phenotyping field discriminating criterion (failed 23 tools) our requirement provide local access. Local installation essential conform privacy security rules. Also, despite some attractive features may seem provide, guarantee efficient pipelines. It also noted 11 never publication 2017 (one 2013, 2014, three 2015, 2016, four 2017), website link broken them (Table largely reflection maintaining resources such activity. 2 summary sources leveraged document type amount relies on, insights need update maintain them.Table 2Data reviewaaZFIN, IMPC, MGD Exomiser. SPIDEX, Pfam, Treefam, GIANT, REACTOME, LRT, InterPro, GWAS, Blast, GO, GERP, dbscSNV, phyloP Xrare. ARIC, GTEx, MetaSVM VARPP. GWAVA DeepPVP.bThe successfully downloaded, installed, study. aZFIN, DeepPVP. bThe Attempts download install Further illustrating problems long-term maintenance software, possible due inaccessible databases, failing dockers, ReadMe In particular, DeepPVP [16.Boudellioua unable follow no phenomenet-vp docker container exists Docker Hub dockerfile recipe provided GitHub repository does build research computing environment containing data; Phenoxome's [42.Wu pull successful there further instructions progression installation; VARPP [11.Anderson required dbNSFP database (version v3.4a) longer possible. finally included downloaded installed next, including brief description rationale algorithms. code analyses https://github.com/whri-phenogenomics/VPSoftware_review. 8.Robinson 9.Cipriani Scholar,45.Smedley al.Next-generation disease-gene Exomiser.Nat. 2004-2015Crossref (197) Java automates contained (and, family members). user-defined applied JANNOVAR [46.Jäger al.Jannovar: java library annotation.Hum. 548-555Crossref (49) functional annotation, minor allele frequency, expected inheritance pattern, amongst others. Each filtered prioritised score rarity pathogenicity, turn combined gene-specific score. latter obtained PhenoDigm algorithm [37.Sifrim calculated semantic similarity between user-provided HPO-encoded annotations diseases, orthologs mouse zebrafish organisms, protein–protein associated neighbours version 13.0.0 (released September 2021) straightforward, accessed Exomiser's pagei. We Bash script create single-sample-analysis-settings.yml starting preset-exome-analysis.yml example per IRD dataset. run command line single-sample (Java 17.0.0; databases 2109; default transcript annotation):java -Xms2g -Xmx4g -jar exomiser-cli-13.0.0.jar –analysis representative sections HTML output sample Figure S1 supplemental online. Tab-separated (tsv) variety relevant (including frequency score, score) processed statistical comparison, described later PhenIX exomes) [44.Zemojtel evaluates ranks predicted thousands OMIM Orphanet 2019). Therefore, did require any additional installation, same way Exomiser, produces similar exploits while replacing option 'hiPhivePrioritiser: {}' 'phenixPrioritiser: {}'. abnormalities) [35.Robinson (LR) framework. Not rank provides estimate post-test probability calculates extent abnormality too) consistent 1.3.4 26 git cloning repositoryii clear 'readthedocs' pagesiii. makes (we 2109). preferred input Phenopacketsiv, open standard, adopted Global Alliance Genomics Healthv, sharing descriptions linked disease, patient, information. Python Phenopacket single-sample-phenopacket.json analysis:java LIRICAL.jar phenopacket -p -e path/to/Exomiser-data-directory -x prefixOfOutputFile --tsv --output-directory path/to/output-directory diagnoses, together rank, probability, LR),

Language: Английский

Automated prioritization of sick newborns for whole genome sequencing using clinical natural language processing and machine learning DOI Creative Commons
Bennet Peterson, Edgar J. Hernández, Charlotte A. Hobbs

et al.

Genome Medicine, Journal Year: 2023, Volume and Issue: 15(1)

Published: March 16, 2023

Rapidly and efficiently identifying critically ill infants for whole genome sequencing (WGS) is a costly challenging task currently performed by scarce, highly trained experts major bottleneck application of WGS in the NICU. There dire need automated means to prioritize patients WGS. Institutional databases electronic health records (EHRs) are logical starting points with undiagnosed Mendelian diseases. We have developed rapid (rWGS WGS) directly from clinical notes. Our approach combines natural language processing (CNLP) workflow machine learning-based prioritization tool named Phenotype Search Engine (MPSE). MPSE accurately robustly identified NICU selected Rady Children's Hospital San Diego (AUC 0.86) University Utah 0.85). In addition effectively WGS, scores also strongly diagnostic cases over non-diagnostic cases, projected yields exceeding 50% throughout first second quartiles score-ranked patients. results indicate that an pipeline selecting acutely neonatal intensive care units (NICU) can meet or exceed obtained through current selection procedures, which require time-consuming manual review notes histories specialized personnel.

Language: Английский

Citations

15
The impact of damaging epilepsy and cardiac genetic variant burden in sudden death in the young DOI Creative Commons
Megan J. Puckelwartz, Lorenzo L. Pesce, Edgar J. Hernández

et al.

Genome Medicine, Journal Year: 2024, Volume and Issue: 16(1)

Published: Jan. 16, 2024

Sudden unexpected death in children is a tragic event. Understanding the genetics of sudden young (SDY) enables family counseling and cascade screening. The objective this study was to characterize genetic variation an SDY cohort using whole genome sequencing.

Language: Английский

Citations

6
Addressing diagnostic gaps and priorities of the global rare diseases community: Recommendations from the IRDiRC diagnostics scientific committee DOI Creative Commons
David R. Adams, Clara D. van Karnebeek, Sergi Beltrán

et al.

European Journal of Medical Genetics, Journal Year: 2024, Volume and Issue: 70, P. 104951 - 104951

Published: June 6, 2024

The International Rare Diseases Research Consortium (IRDiRC) Diagnostic Scientific Committee (DSC) is charged with discussion and contribution to progress on diagnostic aspects of the IRDiRC core mission. Specifically, goals include timely diagnosis, use globally coordinated pipelines, assessing impact rare diseases affected individuals. As part this mission, DSC endeavored create a list research priorities achieve these goals. We present those along current, global disease needs opportunities that support our prioritization. In discussion, we also provide clinical vignettes illustrating real-world examples challenges.

Language: Английский

Citations

6
Current genetic diagnostics in inborn errors of immunity DOI Creative Commons
Sandra von Hardenberg,

Isabel Klefenz,

Doris Steinemann

et al.

Frontiers in Pediatrics, Journal Year: 2024, Volume and Issue: 12

Published: April 10, 2024

New technologies in genetic diagnostics have revolutionized the understanding and management of rare diseases. This review highlights significant advances latest developments inborn errors immunity (IEI), which encompass a diverse group disorders characterized by defects immune system, leading to increased susceptibility infections, autoimmunity, autoinflammatory diseases, allergies, malignancies. Various diagnostic approaches, including targeted gene sequencing panels, whole exome sequencing, genome RNA or proteomics, enabled identification causative variants These not only facilitated accurate diagnosis IEI but also provided valuable insights into underlying molecular mechanisms. Emerging technologies, currently mainly used research, such as optical mapping, single cell application artificial intelligence will allow even more aetiology hereditary near future. The integration clinical practice significantly impacts patient care. Genetic testing enables early diagnosis, facilitating timely interventions personalized treatment strategies. Additionally, establishing is necessary for counselling prognostic assessments. Identifying specific associated with paved way development therapies novel therapeutic approaches. emphasizes challenges related diseases provides future directions, specifically focusing on IEI. Despite tremendous progress achieved over last years, several obstacles remain become important due increasing amount data produced each patient. includes, first foremost, interpretation unknown significance (VUS) known genes (GUS). Although contributed other further exchange between experts from different disciplines, establishment comprehensive guidelines are crucial tackle remaining maximize potential field

Language: Английский

Citations

5
Phenotype-aware prioritisation of rare Mendelian disease variants DOI Creative Commons
Catherine Kelly, Anita Szabó, Nikolas Pontikos

et al.

Trends in Genetics, Journal Year: 2022, Volume and Issue: 38(12), P. 1271 - 1283

Published: Aug. 4, 2022

Next-generation sequencing technologies have made achieving a molecular diagnosis for rare genetic disorder more and feasible and, in turn, enabled personalised clinical management of the affected patients their families.Identifying one or two variants that are responsible certain disease phenotype from millions identified by can be time-consuming expensive.Numerous phenotype-aware variant prioritisation (VP) software tools now exist to help semi-automate process diseases.Although many published VP limitations, show lack maintenance, become soon unfit usage, several up-to-date demonstrate an impressive capacity prioritising diagnoses when tested on real patient data.Adopting diagnostics settings efficiently assist multidisciplinary teams clinicians scientists reporting disease. A analysis data Mendelian diseases has huge impact families. Numerous been developed automate this process, shorten diagnostic odyssey, but performance statistics limited. Here we identify, assess, compare all up-to-date, freely available, programmatically accessible using whole-exome, retinal dataset 134 individuals with diagnosis. All were able identify around two-thirds as top-ranked candidate, LIRICAL performing best overall. Finally, discuss challenges overcome most cases remaining undiagnosed after current, state-of-the-art practices. With approximately 80% having origin, identifying correct causative single-gene disorders creates greater potential informed through precision medicine recommendation drug trials, rather than only treating evident symptoms. Improvements information at scale parallelisation (next-generation sequencing) greatly increased quantities genomic production lower overall costs, shown recent completion 100,000 Genomes Project UK [1.Smedley D. et al.100,000 pilot rare-disease health care - preliminary report.N. Engl. J. Med. 2021; 385: 1868-1880Crossref PubMed Scopus (192) Google Scholar]. Whole-exome (WES) is still commonly used method, exome (~2% human genome) harbours ~85% currently known disease-causing sequence [2.Caspar S.M. al.Clinical sequencing: raw lifetime value.Clin. Genet. 2018; 93: 508-519Crossref (64) The candidate typical WES experiment often derived 60 000 100 affecting protein-coding regions, which nearly will benign unrelated [3.De La Vega F.M. al.Artificial intelligence enables comprehensive genome interpretation nomination diseases.Genome 13: 153Crossref (25) However, filtering review involve tens, if not few hundreds, usually both expensive done via manual scientists. Around one-third children born do live see fifth birthday [4.Wright C.F. al.Paediatric genomics: diagnosing children.Nat. Rev. 19: 253-268Crossref (298) Scholar], so it vital rapid yet, traumatic wait time lengthy (e.g., median 6 years Project) offers possibility disease-causative efficiently, sometimes within minutes. These discard large likely benign, common strategies based publicly available gnomAD) in-house databases. vast majority prioritise single-nucleotide (SNVs) small insertion/deletions (indels) formatted call format (VCF) files. To determine SNVs/indels, incorporate existing silico pathogenicity prediction restrict patients' VCF files interest range methods. They include function-prediction methods MutationTaster, PolyPhen-2, SIFT), likelihood each missense causing pathogenic changes protein structure function; phylogenetic conservation GERP++, phastCons, phyloP), measure degree given nucleotide site; other methods, concern tailored use deep neural networks MVP, PrimateAI); ensemble CADD, DANN, REVEL), integrate multiple component [5.Li al.Performance evaluation pathogenicity-computation variants.Nucleic Acids Res. 46: 7793-7804Crossref (139) Despite availability wide tools, improvements needed discriminate reported specificity 65%; furthermore, sensitivities ranging 51% 96% (median, 88%), relying algorithm-predicted generate number false positive candidates aim automating where relevance gene patient's taken into account, virtually incorporation standardised phenotypic terms, drawing 15 terms Human Phenotype Ontology (HPO) [6.Köhler S. al.The 2021.Nucleic 2020; 49: D1207-D1217Crossref (363) This ultimately significant addition; example, Exomiser (among first its kind) [7.Bone W.P. al.Computational model organism phenotypes improves efficiency.Genet. 2016; 18: 608-617Abstract Full Text PDF (63) Scholar,8.Robinson P.N. al.Improved prioritization genes cross-species comparison.Genome 2014; 24: 340-348Crossref (235) Scholar] demonstrated top diagnosed 20–77% (using variant-based filtering) 96–97% (with addition HPO terms) simulated across different mode inheritances (MOIs) well 3% 74% inferred MOIs [9.Cipriani V. al.An improved phenotype-driven tool prioritization: benchmarking exomiser whole-exome data.Genes. 11: 460Crossref (28) date (different) and/or very datasets, limited comparison. Strikingly, specific always claims outperform relatively literature. Here, set out perform thorough literature tools. Building previous then conducted unbiased comparison selected whole-exomes inherited (IRDs) detailed search was carried list would meet following criteria: (i) directly accepting files; (ii) describe phenotypes; (iii) being (last updated since 2018); (iv) academic use; (v) local, programmatic access (and therefore safer opposed web-based allowing processing scale). Literature searches combination keywords (i.e., 'exome', 'genome', 'variant prioritisation', alternative spelling prioritization', 'human ontology') returned about 400 peer-reviewed journal articles (11 March 2022) (Figure 1). Articles screened those publications involved initially gave 37 Scholar,10.Alemán A. al.A interactive framework studies.Nucleic 42: W88-W93Crossref (35) Scholar, 11.Anderson al.Personalised analytics diagnostics.Nat. Commun. 2019; 10: 5274Crossref (10) 12.Antanaviciute al.OVA: integrating physical biomedical domain ontologies enhanced prioritization.Bioinformatics. 2015; 31: 3822-3829PubMed 13.Bertoldi L. al.QueryOR: web platform prioritization.BMC Bioinform. 2017; 225Crossref (18) 14.Birgmeier al.AMELIE speeds matching genotype primary literature.Sci. Transl. 12eaau9113Crossref (30) 15.Bosio M. al.eDiVA-classification diagnostics.Hum. Mutat. 40: 865-878Crossref (12) 16.Boudellioua I. al.DeepPVP: phenotype-based learning.BMC 20: 65Crossref (36) 17.Boudellioua al.Semantic novel variants.PLoS Comput. Biol. 13e1005500Crossref (24) 18.Chiara al.VINYL: Variant prIoritizatioN survivaL analysis.Bioinformatics. 36: 5590-5599Crossref (4) 19.Desvignes J.-P. al.VarAFT: annotation filtration system next generation data.Nucleic W545-W553Crossref (114) 20.Holt J.M. al.VarSight: prioritizing clinically binary classification algorithms.BMC 1-10Crossref (11) 21.Holtgrewe al.VarFish: DNA research.Nucleic 48: W162Crossref 22.Hombach al.MutationDistiller: user-driven identification 47: W114-W120Crossref 23.Hunt S.E. al.Annotating Ensembl Effect Predictor-a tutorial.Hum. 43: 986-997Crossref 24.Ip E. al.VPOT: customizable ordering annotated variants.Genom. Proteom. 17: 540Crossref (7) 25.James R.A. visual curatorial approach discovery genome-wide diagnostics.Genome 8: 13Crossref 26.Javed al.Phen-Gen: combining analyze disorders.Nat. Methods. 935-937Crossref (106) 27.Kennedy B. al.Using VAAST disease-associated next-generation data.Curr. Protoc. Hum. 81: 6PubMed 28.Koile al.GenIO: phenotype-genotype server genomics diseases.BMC 25Crossref (9) 29.Li M.J. al.wKGGSeq: strategy-based disease-targeted online facilitate studies disorders.Hum. 496-503Crossref (8) 30.Li Q. al.Xrare: machine learning method jointly modeling evidence diagnosis.Genet. 21: 2126-2134Abstract (37) 31.Li Z. al.PhenoPro: toolkit assisting disease.Bioinformatics. 35: 3559-3566Crossref (21) 32.Manshaei R. al.GeneTerpret: multilayer interpretation.BMC 15: 31Google 33.Muller H. al.VCF.Filter: disease-linked 45: W567-W572Crossref (19) 34.O'Brien T.D. (AI)-assisted reanalysis aids new reduces laboratory.Genet. 2022; 192-200Abstract 35.Robinson al.Interpretable ratio paradigm.Am. 107: 403-417Abstract 36.Seo G.H. al.Diagnostic yield utility whole automated system, EVIDENCE.Clin. 98: 562-570Crossref (51) 37.Sifrim al.eXtasy: fusion.Nat. 2013; 1083-1084Crossref (131) 38.Singleton M.V. al.Phevor combines accurate alleles single nuclear families.Am. 94: 599-610Abstract (137) 39.Stelzer G. al.VarElect: variation prioritizer GeneCards Suite.BMC Genomics. 444Crossref (126) 40.Trakadis Y.J. al.PhenoVar: polymalformative syndromes.BMC 7: 22Google 41.Ward al.Clin.iobio: collaborative workflow enable team-based genomics.J. Pers. 12: 73Crossref (1) 42.Wu C. al.Rapid exomes Phenoxome: computational approach.Eur. 27: 612-620Crossref (13) 43.Yang Wang K. Genomic ANNOVAR wANNOVAR.Nat. 1556-1566Crossref (566) 44.Zemojtel T. al.Effective genome.Sci. 6252ra123Crossref (188) prune according aforementioned criteria. Remarkably, seven passed five criteria final testing Table 1 shows details retrieved corresponding selection process.Table 1Selection suitability criteriaaTable criteriaaaA grey cell indicates feature present.bFollowing review, viable met terms; last available; access) aA present. bFollowing Most (33) accept file, standard file storing data. total 28 'phenotype-aware' simply prioritisation; they allow integrative HPO, de facto phenotyping field discriminating criterion (failed 23 tools) our requirement provide local access. Local installation essential conform privacy security rules. Also, despite some attractive features may seem provide, guarantee efficient pipelines. It also noted 11 never publication 2017 (one 2013, 2014, three 2015, 2016, four 2017), website link broken them (Table largely reflection maintaining resources such activity. 2 summary sources leveraged document type amount relies on, insights need update maintain them.Table 2Data reviewaaZFIN, IMPC, MGD Exomiser. SPIDEX, Pfam, Treefam, GIANT, REACTOME, LRT, InterPro, GWAS, Blast, GO, GERP, dbscSNV, phyloP Xrare. ARIC, GTEx, MetaSVM VARPP. GWAVA DeepPVP.bThe successfully downloaded, installed, study. aZFIN, DeepPVP. bThe Attempts download install Further illustrating problems long-term maintenance software, possible due inaccessible databases, failing dockers, ReadMe In particular, DeepPVP [16.Boudellioua unable follow no phenomenet-vp docker container exists Docker Hub dockerfile recipe provided GitHub repository does build research computing environment containing data; Phenoxome's [42.Wu pull successful there further instructions progression installation; VARPP [11.Anderson required dbNSFP database (version v3.4a) longer possible. finally included downloaded installed next, including brief description rationale algorithms. code analyses https://github.com/whri-phenogenomics/VPSoftware_review. 8.Robinson 9.Cipriani Scholar,45.Smedley al.Next-generation disease-gene Exomiser.Nat. 2004-2015Crossref (197) Java automates contained (and, family members). user-defined applied JANNOVAR [46.Jäger al.Jannovar: java library annotation.Hum. 548-555Crossref (49) functional annotation, minor allele frequency, expected inheritance pattern, amongst others. Each filtered prioritised score rarity pathogenicity, turn combined gene-specific score. latter obtained PhenoDigm algorithm [37.Sifrim calculated semantic similarity between user-provided HPO-encoded annotations diseases, orthologs mouse zebrafish organisms, protein–protein associated neighbours version 13.0.0 (released September 2021) straightforward, accessed Exomiser's pagei. We Bash script create single-sample-analysis-settings.yml starting preset-exome-analysis.yml example per IRD dataset. run command line single-sample (Java 17.0.0; databases 2109; default transcript annotation):java -Xms2g -Xmx4g -jar exomiser-cli-13.0.0.jar –analysis representative sections HTML output sample Figure S1 supplemental online. Tab-separated (tsv) variety relevant (including frequency score, score) processed statistical comparison, described later PhenIX exomes) [44.Zemojtel evaluates ranks predicted thousands OMIM Orphanet 2019). Therefore, did require any additional installation, same way Exomiser, produces similar exploits while replacing option 'hiPhivePrioritiser: {}' 'phenixPrioritiser: {}'. abnormalities) [35.Robinson (LR) framework. Not rank provides estimate post-test probability calculates extent abnormality too) consistent 1.3.4 26 git cloning repositoryii clear 'readthedocs' pagesiii. makes (we 2109). preferred input Phenopacketsiv, open standard, adopted Global Alliance Genomics Healthv, sharing descriptions linked disease, patient, information. Python Phenopacket single-sample-phenopacket.json analysis:java LIRICAL.jar phenopacket -p -e path/to/Exomiser-data-directory -x prefixOfOutputFile --tsv --output-directory path/to/output-directory diagnoses, together rank, probability, LR),

Language: Английский

Citations

21