Semiautomated approach focused on new genomic information results in time and effort- efficient reannotation of negative exome data

doi:10.21203/rs.3.rs-3146571/v1

Download PDF

Research Article

Semiautomated approach focused on new genomic information results in time and effort- efficient reannotation of negative exome data

https://doi.org/10.21203/rs.3.rs-3146571/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 27 Mar, 2024

Read the published version in Human Genetics →

You are reading this latest preprint version

Introduction:

Most rare disease patients (75 − 50%) undergoing genomic sequencing remain unsolved, often due to lack of information about variants identified. Data review over time can leverage novel information regarding disease-causing variants and genes, increasing this diagnostic yield. However, time and resource constraints have limited reanalysis of genetic data in clinical laboratories setting. We developed RENEW, (REannotation of NEgative WES/WGS) an automated reannotation procedure that uses relevant new information in on-line genetic databases to enable rapid review of genetic findings.

Methods

We tested RENEW in an unselected cohort of 1,066 undiagnosed cases from the Mayo Clinic Center for Individualized Medicine using new information in ClinVar, HGMD and OMIM between the date of previous analysis/testing and April of 2022.

Results

5,741 variants prioritized by RENEW were rapidly reviewed by variant interpretation specialists. Mean analysis time was approximately 20 seconds per variant (32 hours total time). Reviewed cases were classified as: 879 (93.0%) undiagnosed, 63 (6.6%) putatively diagnosed, and 4 (0.4%) definitively diagnosed.

Discussion

New strategies are needed to enable efficient review of genomic findings in unsolved cases. We report on a fast and practical approach to address this need and improve overall diagnostic success in patient testing through a recurrent reannotation process.

Next generation sequencing has transformed the clinical genetics practice providing a cost-effective and comprehensive approach to detect variants in multiple genes concomitantly.(Ashley 2016) The diagnostic yield of exome sequencing (ES) in clinical settings varies between 25–50%, depending on factors including the underlying condition, previous genetic testing, and inherent limitations to the testing.(Lee 2014, Yang 2014, Farwell 2015, Posey 2017) In many instances, variants in disease-causing genes are identified which lack sufficient data to support a clear interpretation resulting in a variant of uncertain significance (VUS) classification according to the American College of Medical Genetics and Genomics/Association for Molecular Pathology (ACMG/AMP) guidelines.(Richards 2015) Similarly, many genes lack a clear disease association and are classified as genes of uncertain significance (GUS).

Tremendous growth in the understanding of the genetic etiology of disease is occurring. This is the result of increasing use of genetic testing across many clinical disciplines, and the corresponding increase in research programs to functionally characterize findings of uncertain significance and establish new gene-phenotype associations.(Klee 2021) Additionally, the wide adoption of collaborative sharing of genetic data worldwide has also fueled novel disease-phenotype-gene discovery and characterization(Ramoni 2017), and new knowledge is continuously added to genetic databases, including new and reclassified disease-causing variants and new gene-disease associations. As a result, revisiting non-diagnostic genetic test results over time can increase the diagnostic yield by up to 10%, depending on the disease and reported phenotypes afflicting the patients. (Wenger 2017, Ewans 2018, Nambot 2018, Epilepsy Genetics 2019, Liu 2019, Salfati 2019) Despite the obvious benefit, systematic reinterpretation of genomic data can be costly, creating a need for tools or systems to streamline this process and facilitate rapid data review, to ensure patients are not missing out on emergent answers to their genetic disease. Given the ever-growing number of individuals sequenced,(O'Daniel 2017, Ji 2021) this missed opportunity is likely to increase with time and creates the need for new strategies for reviewing existing data in an efficient manner.

Several approaches to automate and facilitate data reanalysis have been described that rely in a wide range of strategies (e.g. tailored scoring systems or automated review of the scientific literature) with different levels of success. (Baker 2019, Ji 2021, Seo 2022) Nevertheless, the complexity and subjectivity of genetic data interpretation may prevent full automated reinterpretation of genomic data. For this reason, we developed “REannotation of NEgative WES/WGS” (RENEW), a novel semi-automated strategy leveraging knowledge from three main genomic databases [ClinVar, Human Gene Mutation Database Professional® (HGMD), and OMIM] (Stenson 2003, Landrum 2018) to clarify ambiguous genetic findings and identify new findings in unsolved cases. RENEW downloads and normalizes information from new releases of these databases and extracts new and updated information by comparing it with the previous release. Variant annotations are added and filtered for results of high interest. Expert variant curators assess the information in an expedited fashion, enabling a high number of cases to be reviewed in a matter of minutes, and critical new findings identified. This strategy provides a cost-effective approach to systematically reanalyze genomic findings, providing evidence-based, genetic diagnoses in patients with previously ambiguous or negative test results.

Patient sequencing data

Patients were identified through the Mayo Clinic Department of Clinical Genomics and evaluated as described in Klee at al. 2021.(Klee 2021) All individuals underwent ES in CLIA-certified laboratories as part of their clinical care, and test results were considered by the clinical team as uninformative or insufficient to explain fully the patient’s phenotype. Patients were otherwise unselected regarding phenotype, age, sex, race or any other demographic characteristic. Informed consent using a protocol approved by our institutional review board was obtained from all patients or patient representatives. ES raw data was processed through our in-house developed bioinformatic variant calling and annotation pipeline. (Asmann 2012) All cases were classified by expected mode of inheritance (MOI) [autosomal dominant inherited (AD), autosomal recessive/de novo dominant (AR/DND) or Unknown] according to the phenotype observed, family history, and affected status following the decision tree described in Fig. 1. If pertinent information was not available, MOI was treated as “UNKNOWN”. When possible, “de novo” status for all variants was assigned.

Twenty-five cases for which the molecular diagnosis was identified during the time between the release of an uninformative clinical report and the RENEW analysis described herein were selected as positive control test cases. In all cases, the new information had been submitted to ClinVar, HGMD and/or OMIM. These cases were similarly unselected for any demographic or phenotype.

Reannotation of sequencing data

Data from the public genetic databases ClinVar, HGMD, and OMIM were downloaded at regular intervals and used to identify new or modified entries from the previous download. These differential annotations were applied to genomic test results from cases registered in the RENEW process. The heterogeneity of formats used in genetic databases required data ingestion and transformation prior to differential analysis. This information was integrated into Mayo Clinic’s proprietary BioR Catalog using a TJSON format, (Kocher 2014) enabling streamlined comparison of data from all annotation sources. Clinical test results were annotated with the corresponding Ann-Δ. Variant-level annotations were matched by genomic position, reference allele, and alternate allele. Gene-level annotations were matched by gene genomic coordinates. A command line version of the BioR Toolkit (Kocher 2014) was used for this step.

Annotations were applied for all differential findings between the current RENEW review date and either the previous RENEW review date for cases existing in the system ("prior-cases"), or date of the original test results for new cases enrolled in the RENEW process (“new-cases”). This change in time between genomic results review was “annotation delta” (Ann-Δ). For the example cohort used in this manuscript Ann-Δ was determined for each case since the corresponding baseline to April of 2022, which was 6 months for prior-cases.

Variant Filtering and Prioritization Methods

Variants were filtered to remove results due to uninformative or spurious database changes (e.g. punctuation changes vs new phenotype described), and prioritized clinically relevant changes (e.g. new gene-phenotype association vs variants classified as benign).

General filters were applied to remove variants located in regions defined by RepeatMasker(AFA 2015). Variants were filtered based on gnomAD(Karczewski 2020) minor allele frequency as described in Table 1, and variants were removed that were prioritized in more than 5% of the RENEW cases. Variants classified as “low impact” by CAVA (clinical annotation of variants) (Munz 2015) prediction were filtered. In addition, variants were filtered that did not new or modified differential annotation that include one of the following: (i) HGMD entries classified as likely disease-causing mutations or disease-causing mutation, (ii) ClinVar entries classified as likely pathogenic or pathogenic, (iii) found in an OMIM gene with changes impacting inheritance and phenotype description. Additionally, case data was filtered as described in Table 1 using sample set [trio (proband and both parents), singleton, or other], parental affected status (affected or unaffected) and de novo variant status (yes or no).

Table 1

Query-specific filters applied for variant prioritization through RENEW.
Query	Database	Sample set	Gene MOI described in OMIM	gnomAD threshold	Parents affected status	Variant De Novo status	MOI indicated	Changes in database detected	Additional filters
ClinVar_De novo_Dominant_X linked	ClinVar	Not used	AD, X-linked, None	< 0.1%	Not Used	Not used	AD, DND, X	- New entries or old entries with new classification of: • Likely Pathogenic • Pathogenic • Likely Pathogenic/Pathogenic
Clinvar_Recessive	ClinVar		AR, None	< 1%			AR		-Variants in gene equal or above 2
HGMD	HGMD		Not used	< 1%			Not used	New entries or updated entries with DM and DM?
OMIM_Singleton_Denovo_Dominant	OMIM	Singleton, other		< 0.1%	Both parents unaffected		DND	-New phenotype added -New AD inheritance added
OMIM_Singleton_Dominant				< 0.1%	At least one parent affected		AD	-New phenotype added -New AD inheritance added
OMIM_Singleton Recessive				< 1%	Both parents unaffected		AR	-New phenotype added -New AR inheritance added	-Variants in gene equal or above 2
OMIM_Singleton_X-linked				< 1%	Not used		X	-New phenotype added -New XLR inheritance added	- Variant is in chromosome X
OMIM_Trio_DeNovo		Trio, other		< 1%	Both parents unaffected	Yes	DND	-New phenotype added -New AD inheritance added
OMIM_Trio_Dominant				< 0.1%	At least one parent is affected, and the variant is present in affected parent	No	AD	-New phenotype added -New AD inheritance added
OMIM_Trio_Recessive				< 1%	Both parents unaffected	No	AR	-New phenotype added -New AR inheritance added	-Variants in gene equal or above 2
OMIM_TRIO_X-linked				< 1%	Not used	Yes or present only in mother	X	-New phenotype added -New XLR inheritance added	- Variant is in chromosome X

Manual review of prioritized variants

Seven variant interpretation experts within the Translational Omics Program in the Mayo Clinic Center for Individualized Medicine (Klee 2021) reviewed variants prioritized by RENEW and tracked the analysis time per case while categorizing the cases as: 1) No relevant findings, 2) variant requiring further study, and 3) putative likely-pathogenic or pathogenic finding. To streamline review, results are presented with case phenotype, genomic position of the variant, reference and alternate alleles, genomic position genotype of all samples included in RENEW, HUGO gene symbol, cDNA and protein change, OMIM reported phenotype, MOI for the associated gene, and links to the public databases containing relevant information. Database entries were color coded as follows: black for unchanged text, blue for new information added, and red for information deleted. For results arising from ClinVar and HGMD (variant-specific resources), the total number of case-specific rare variants (gnomAD AF < 1%) identified in the target gene was provided, to identify possible second-hits for cases with a suspected recessive MOI.

Cohort

1066 unsolved cases were analyzed using the RENEW strategy, consisting of 47.9% female with median age at the time of testing of 14 years of age (range: 0 to 78 years). The cohort included 643 cases with trio sequencing, 273 cases sequenced as singletons, and 150 cases sequenced with other sample configurations (“others”). Median time lapse between the original clinical report and RENEW analysis was 33 months (range: 6 to 141 months – Supplementary Fig. 1). In total, 684 cases had been previously analyzed using the RENEW approach 6 months prior (prior cases) and 382 were analyzed for the first time in RENEW (median baseline: 28 months, range: 6 to 141 months - new cases).

Validation of RENEW approach with independently solved cases

We selected 25 cases with known diagnostic findings following an initial unrevealing clinical report that were used to test the RENEW process. The correct diagnostic finding was prioritized by the RENEW process in all 25 cases. Pertinent annotations were flagged from ClinVar in 11 cases (44%), HGMD in 8 cases (32%) and OMIM in 4 cases (16%). For 2 cases, the information was flagged in all three databases (Table 2).

Table 2

List of cases used to test RENEW’s sensitivity. All cases were solved by our group and new information was made available in genomic databases after the original uninformative clinical report. This new information was captured by the RENEW pipeline for all cases.
Clinical Interpretation Date	Time in RENEW (Months)	New diagnostic gene	Database source	Sample set configuration	Mode of Inheritance	Main Phenotypes
2013-10-23	101	GABBR2	OMIM	Singleton	AR, DND, X	Global developmental delay Muscle hypotonia
2015-04-01	84	FAR1	ClinVar	Trio	AR, DND, X	Spastic paraplegia Cataracts Epilepsy Global developmental delay
2016-05-27	70	CASR	ClinVar	Singleton	AR, DND, X	Difficulty with speech and language Idiopathic basal ganglia calcifications Diminished capacity for concentration
2017-01-27	62	SPEN	ClinVar	Trio	AR, DND, X	Global developmental delay Hypotonia Speech delay Feeding difficulties History of failure to thrive Dysmorphic features Gastroesophageal reflux
2017-05-26	58	CACNA1E	ClinVar	Trio	AR, DND, X	Global developmental delay Epilepsy
2017-05-26	58	NAA15	OMIM	other	AR, DND, X	Developmental delay Gastroesophageal reflux Attention deficit hyperactivity disorder
2017-05-26	58	ASH1L	OMIM	Trio	AR, DND, X	Global developmental delay Epilepsy
2017-06-05	58	STX1B	HGMD	Singleton	Unknown	Lupus PE Stroke Rheumatoid Arthritis Encephalopathy Celiac DM1
2017-10-06	54	CACNA1E	ClinVar	Trio	AR, DND, X	Profound developmental delay Seizures Infantile spasms Myoclonus Suspected opsoclonus
2017-11-03	53	SPTB	OMIM	Singleton	Unknown	Hereditary spherocytosis Status post splenectomy Recurrent fevers of unknown etiology Migraine headaches NOD2 mutations of uncertain significance
2017-11-21	53	DEGS1	HGMD	Singleton	AR, DND, X	Nystagmus Dystonia Global developmental delay Hypotonia
2017-11-24	52	SLC6A19	ClinVar	Trio	AR, DND, X	Short stature Global developmental delay Hyperoxaluria Dysmorphic features Kidney disease
2018-02-07	50	GFAP & ACVRL1	ClinVar	other	AR, DND, X	Leukodystrophy Alexander's disease
2018-03-13	49	MORC2	ClinVar	Trio	Unknown	Ataxic Gait Hereditary Optic Atrophy Global developmental delay
2018-03-23	48	NF1	ClinVar	Trio	Unknown	Global developmental delay
2018-04-26	47	RPS6KA3 & USP9X	HGMD	Trio	AR, DND, X	Autism spectrum disorder Significant speech delay with history of regression Attention deficit hyperactivity disorder Hypertrichosis History of spells and abnormal EEG Epilepsy
2018-04-27	47	ZC4H2	ClinVar HGMD OMIM	Trio	AR, DND, X	Prematurity Arthrogryposis Dilated renal pelvises
2018-05-01	47	COG5 & ALG3	HGMD	Singleton	AR, DND, X	Multiple fetal anomalies Fetus with skeletal dysplasia Abnormal placenta
2018-05-04	47	CSNK2B	ClinVar HGMD OMIM	Trio	AR, DND, X	Encephalopathy Nonverbal learning disability Executive function weaknesses Fine motor dexterity impairment, bilateral Social anxiety Depression
2018-06-01	46	RINT1	HGMD	Trio	AR, DND, X	Liver disease Hepatomegaly
2018-06-04	46	HCFC1	HGMD	Trio	AR, DND, X	Global developmental delay Hypotonia Apraxia of speech Attention deficit hyperactivity disorder Obstructive sleep apnea
2018-08-22	44	UBE3A	HGMD	Trio	AR, DND, X	Abnormal brain MRI suggesting inflammatory changes Epilepsy Speech expression difficulty
2019-04-17	36	GNB1	ClinVar	Singleton	Unknown	Global developmental delay Hypotonia Muscle Epilepsy
2020-04-10	24	PGM1	HGMD	Singleton	AR, DND, X	Congenital Glycosylation Disorder (HCC) Rhabdomyolysis Hypoglycemia Myopathy Congenital (HCC)
2020-12-31	15	TPO	ClinVar	Trio	AR, DND, X	Hypothyroidism Congenital Global developmental delay Hearing Sensorineural hearing loss

Variant filtering and prioritization

For “prior cases” (analyzed 6 months prior) differential data from three databases between November of 2021 to April of 2022 were analyzed. The total number of entries identified included 1,350,315 from ClinVar, 20,266 from HGMD and 1,551 from OMIM, that were subsequently filtered to 33,172, (2.4% of total changes) 8,802 (43.4%) and 59 (3%) respectively (Fig. 2A to C). For “new cases” differential data were included to cover date ranges from the original clinical report (variable per case) to April of 2022, and identified 5,741 new variants of interest. These were significantly enriched (Fig. 3A, p < 0.001) in “new cases” (2801 variants in 382 cases – average of 7.3 variants per case) compared to “prior cases” (2,940 variants in 684 cases – average of 4.3 variants per case). In line with these results, cases lacking any variants of interest were significatively (p < 0.001) less frequent within “new cases” (14 cases, 3.6%) than “prior cases” (107 cases, 15.64%). Time elapsed between the release of the clinical report and RENEW analysis was not directly correlated with higher number of variants identified (R² = 0.1621, Fig. 3B).

Each database resulted in significantly different number of variants highlighted with averages per case of 0.7, 1.68 and 2.99 variants flagged by ClinVar, HGMD, and OMIM respectively (Fig. 3C). For both prior and new cases, cases with trio data presented a significantly lower (P < 0.001) number of variants compared to singletons, (Fig. 3D) while MOI did not significantly impact the number of variants flagged per case (Fig. 3E).

Manual review of prioritized variants

Manual review of variants of interest with RENEW took 32.06 hours (5.34 hours per person) by the expert variant curators, averaging 20.1 seconds per variant. No difference in the average time per variant was found between prior and new cases, sample set, or MOI (Fig. 4A-C).

945 cases had at least one variant of interest for review, with 879 cases (93.0%) remaining unsolved, 63 (6.6%) with findings requiring additional studies to clarify pathogenicity, and 4 cases (0.4%) receiving a putative new diagnosis (Fig. 4D). Cases that remained unsolved needed significantly less time to review compared to cases flagged for further follow up (p < 0.001) and new diagnoses (p = 0.0016) (Fig. 4E). No differences were seen in results when accounting for time since first clinical analysis (Fig. 4F).

RENEW process examples:

To illustrate the RENEW functionality we report on 3 cases involving two genes (AFG3L2 and ZBTB7A) that resulted in a newly solved case that remained unsolved for one year, and the identification of a variant of interest that prompted further research analysis on a case from a patient in a diagnostic odyssey for more than 5 years. Figure 5 describes the timeline of key events leading to the initial lack of genetic findings and subsequent identification of relevant variants regarding gene AFG3L2.

Example: Emerging AFG3L2 disease association

The RENEW process identified a newly described disease-gene relationship in AFG3L2 that led to diagnoses of two patients.

Patient 1: A female patient was referred to the Department of Clinical Genomics at the Mayo Clinic at the age of 8 years for evaluation for optic atrophy. She presented with bilateral vision loss (Rt: 20/160, Lt: 20/200) and was first noted to have vision impairment at 5 years of age. Her 39-year-old father was reported as being blind and was also diagnosed with optic atrophy. Upon evaluation, there was no indication of an intellectual impairment or ataxic gait, though some mild clumsiness and weakness was noted in the proband, while there was no gait concern in her father. She was born at term without complication to consanguineous parents.

Previous targeted genetic testing of MFN2, OPA1, OPA3, TIMM8A and TMEM126A was unrevealing. Mitochondrial testing revealed a homoplasmic variant m.4350C > T also homoplasmic in the unaffected mother and therefore deemed not relevant. The ES clinical lab report, issued in January 2020, reported no variants. The genetic data was initially reviewed by RENEW in October 2020 as a “new case”, and therefore the Ann-Δ was calculated between January and October of 2020 identifying 8 unique, potentially relevant variants within the OMIM and HGMD queries that were determined as not relevant. The case was reclassified as “Nothing relevant found” and remained unsolved. The case was reviewed again in April 2021 as an “old case” with an Ann-Δ calculated between October 2020 and April of 2021. At this time, 1 variant in AFG3L2 was identified through the HGMD query (NM_006796.3: c.1064C > T p.(Thr255Met)), and noted to be heterozygous in proband and father.

A detailed review of the variant and case level data as previously described, (Klee 2021) revealed the finding to be relevant to the patient’s phenotype. In brief, NM_006796.3:c.1064C > T p.(Thr255Met) was reported in an individual that presented with early onset optic atrophy, mild motor disturbance and diffuse mild cerebral atrophy in adolescence(Caporali 2020) and was found to segregate with disease in other relatives. The authors noted an expansion of the AFG3L2-associated disease, with pathogenic missense variants impacting the ATPase domain being causative for optic atrophy and incomplete penetrance of mild ataxia (MIM: 618977).(Caporali 2020) The manuscript describing these findings was published in July of 2020 and the HGMD entry for the same variant was updated in January of 2021, being captured by RENEW at this point. After revision by the medical provider, this information was relayed to the clinical lab which generated an amended report including the variant and classifying it as likely pathogenic.

Patient 2: A 11-year-old male was referred for genetic testing at the age of 6 years for severe global developmental delay, failure to thrive, and tonic seizures consistent with Lennox-Gastaut syndrome, with an age of onset at 6 months. An initial brain MRI noted thinning of the corpus callosum and delayed myelination. The clinical ES test in 2018 reported a de novo variant in AFG3L2 (NM_006796.3:c.1219G > A p.(Asp407Asn)), classified as VUS. At this time, the variant was determined to not be causative for the patient’s condition and therefore the case was unsolved. Reanalysis of the case using the RENEW approach in April 2021 identified an annotation arising from the same manuscript described for Patient 1 described above. The AFG3L2 variant was associated within an OMIM entry linked to a new gene-disease association with Optic Atrophy 12 (MIM:618977). The patient variant was novel and consequently the ClinVar and HGMD queries (variant-specific) did not highlight any significant findings. Given the typical adolescent age of onset for this disease and the de novo status of this variant, the patient’s medical records were reviewed, and a recent brain MRI noted thinning of the optic chiasm and mild atrophy of the optic nerves. There has not been an indication of visual impairment of the child to date. A variant occurring in the same codon (NM_006796.3: c.1220A > G p.(Asp407Gly)) was previously reported in a patient with optic atrophy and determined to impact protein function. (Caporali 2020) Upon request, the clinical lab re-classified this variant as Likely Pathogenic after applying the ACMG rules PS2, PM2_supporting, PM5, and PP3. While this finding does not provide a complete genetic explanation to the patient’s phenotypic presentation, it does explain part of it and significantly improved the patient’s individualized medical care. This case is still being reviewed through RENEW for future additional findings that might explain the phenotype of the patient fully.

Example

Identification of ZBTB7A variant prompting research analysis.

A male patient currently 14 years old was seen at our institution at 6 months of age due to developmental delays, macroglossia, and recurrent upper respiratory tract infections since birth, requiring hospitalization due to heavy congestion and severe coughing spells. He was born to a G2P2 mother at 39 weeks’ gestation (4.86 kg, 51 cm) and noted to be macrocephalic (97th %). He was noted to have tonsillar hypertrophy with no indication of glossoptosis for which he underwent a tonsillectomy and repeat adenoidectomies due to regrowth. He required continuous positive airway pressure (CPAP) while sleeping and underwent bilateral myringotomy with placement of PE tubes for recurrent ear infection. No facial dysmorphism was noted. He was also diagnosed with attention-deficit/hyperactivity disorder (ADHD) at 7 years of age. Extensive biochemical and genetic testing was performed and was unrevealing. Specifically, TSH, amino acids, acylcarnitine, total and free carnitine, and alpha-glucosidase were all within normal levels. Family history was significant for both parents having developmental delay and intellectual disability. His father was also noted to be macrocephalic while his mother was diagnosed with ADHD and asthma as a child. Additionally, there are several maternal family members, including both great aunts and uncles, as well as their children, who reportedly suffer from learning disabilities, intellectual disability, macrocephaly, and macroglossia. The proband’s older, full brother shared a similar clinical presentation, though more mildly affected, which included a BMI of > 95%, diagnosis of attention deficit disorder (ADD), mild to moderate developmental, motor and speech delays, macrocephaly, and urinary incontinence. He had a history of obstructive sleep apnea that resolved with adenoidectomy and tonsillectomy. He was also noted to have mild mandibular hypoplasia and has a history of chronic bilateral otitis media. Subsequently, clinical ES was completed on the proband at 8 years of age and included his similarly affected brother for segregation analysis. A VUS in RASA1 (NM_002890:c.407C > A p.(Pro136His)) was reported but discarded after internal review as it was not present in the similarly affected brother.

RENEW analysis, performed 5 years after the clinical report, identified 5 candidate variants from the 3 queries: ClinVar (1), HGMD (1), OMIM (3). Four of the variants were discarded since they did not constitute a good phenotypic match with the patient, whereas a novel nonsense variant impacting ZBTB7A was highlighted in the OMIM query due to the annotation of a new gene-disease association (MIM: 619769) with an AD inheritance. The variant (NM_015898.4: c.650T > G p.(Leu217Ter)) was found to be heterozygous in both the proband and brother’s DNA, and is likely maternally-inherited given the significant family history. This variant occurs in exon 2 of a 3-exon gene and is predicted to elicit nonsense-mediated decay of the truncated mRNA. The phenotype associated with ZBTB7A (Ohishi 2020, von der Lippe 2022) (Table 3) presents a significant overlap to the one seen in this family including macrocephaly, persistent hypertrophy of both the adenoid and tonsillar tissue with recurrent respiratory infections and otitis media, sleep apnea, cognitive impairment with motor and speech delays, mild hypotonia, autistic features and a high BMI > 95% for their respective ages. Furthermore, given the evidence that ZBTB7A transcriptionally represses fetal hemoglobin (HbF), (Masuda 2016) clinical testing of HbF was pursued indicating significantly increased levels in both the proband and the affected brother (8.4 and 4% respectively), consistent with previously described affected individuals. (Ohishi 2020, von der Lippe 2022)

Table 3

Comparison between described phenotype associated with *ZBTB7A* and the family identified by the RENEW method.
Feature	Charlotte von der Lippe et al (2022) Ohishi et al (2020)	Proband	Brother
ZBTB7A-variant Nonsense/Frameshift	8/12	c.650T > G. p.(Leu217Ter)	c.650T > G. p.(Leu217Ter)
ZBTB7A-variant Missense	5/13	N/A	N/A
Lymphoid hypertrophy/adenoid overgrowth	12/12	Yes	Yes
Cognitive impairment/ID	13/13	Yes	Yes
Motor millstones delayed	10/12	Yes	Yes
Speech delay	10/13	Yes	Yes
Macrocephaly (≥ 90p)	12/13	Yes	Yes
Weight > 97th percentile	8/12	Yes	Yes
Frequent upper airway infections	10/11	Yes	Yes
Sleep apnea	10/12	Yes	Yes
Autistic features	7/12	Yes	Yes
Elevated HbF	5/6	Yes	Yes

The clinical laboratory was approached with this new information and agreed to include this new variant in an edited clinical report. Although the few evidence available at this point resulted in a VUS classification, the high phenotypic overlap and deleterious nature of the variant prompted the selection of this case for further research functional validation, currently underway.

Exome and genome clinical testing have led to genetic diagnoses for many patients who were on a proverbial diagnostic odyssey. Despite this, the majority of patients receiving what can be considered the most comprehensive germline test available remain undiagnosed. (Lee 2014, Yang 2014, Farwell 2015, Posey 2017) Our incomplete understanding of the genetics of disease to explain phenotypes presented by a patient at the time of testing inhibits interpretation of most variants identified when sequencing a patient. At the same time, clinical genetics is experiencing tremendous growth in understanding the etiology of disease, driven by widespread patient testing. New genetic findings are continuously being indexed in the literature and submitted to structured databases creating the opportunity that previously unexplained cases may be resolved with reannotation and analysis. Despite the obvious benefit of genetic test reanalysis for patients with unsolved disease, high costs, a paucity of methods, and limited resources have prevented wide-spread adoption of systematic genomic data review. We implemented the RENEW process to address this unmet need, providing an efficient strategy for automated reannotation and variant prioritization predicated on new data in ClinVar, HGMD, and OMIM. This approach provides information both at the variant and gene level that allows for efficient, systematic, and continuous review of unsolved cases.

The large number of new findings and subsequent database changes observed with each release of ClinVar, HGMD, and OMIM creates emergent opportunities for patient diagnoses. However, the high number of clinically irrelevant changes found in each database release creates a technical burden that limits the utility of any approach using unfiltered differential analyses. As an illustrative example, when evaluated the changes found in two OMIM releases (October 2021 to April 2022, Fig. 2C), 117 entries were added and 1,434 entries were modified, resulting in 9.7% of all database entries changed. Scrutiny of these changes reveals many modified entries which have no clinical significance, such as punctuation changes or edits that do not impact the variant/gene classification. It is therefore necessary to remove clinically irrelevant changes from the differential data. Furthermore, case-specific prioritization of the resulting variants is also needed to prioritize new information that may impact interpretation of the variants in light of the underlying case presentation. To illustrate this, of the 1,551 database entry changes noted in OMIM above, clinically meaningful information was only found in 59 genes. After using case-specific prioritization these entries translated into an average of 3 variants per case selected for review. In the absence of full reinterpretation automation, RENEW implements key steps to reduce the burden of systematic review by eliminating irrelevant changes and prioritizing new information that is most likely to lead to a clinical diagnosis.

We evaluated RENEW with 25 case examples with an uninformative ES clinical report and subsequent clinical diagnostic result identified through in-depth manual review and research studies by our group (Table 2). The RENEW process successfully prioritized the causal genetic findings for all cases, independent of the sample set configuration or MOI expected for these cases. Neurological phenotypes were predominant in these cases. However, as RENEW does not use phenotypic information for variant filtering or prioritization, process performance should not be impacted by the phenotypic makeup of the test cases.

In a prospective analysis of ongoing clinical translational activities, RENEW processed a cohort of 1,066 unsolved cases, calculating the Ann-Δ for each specific case starting from the date of the original clinical report (new cases) or the date of the last RENEW analysis (6 months prior), (prior cases) to April 2022. 5,741 variants in 945 cases were selected for review and presented in a structured report including case relevant information. Seven expert variant scientists completed the analysis in ~ 32 hours, with an average of 5.34 hours per person and 20.1 seconds per variant. The prospective study at scale, illustrates the highly efficient strategy for reviewing large case cohorts for putative new diagnostic findings and suggests this approach can be systematically used at regular intervals. The prospective cohort analysis with RENEW resulted in 4 (0.4%) cases with a putative new diagnosis and high value variant finding for 63 (6.6%) additional cases that were prioritized for additional study (Fig. 4D to F).

A higher number of variants were identified for review in new cases compared to prior cases (Fig. 3A), likely due to new cases having a larger Ann-Δ. However, there was no direct correlation between time elapsed since the original clinical report and number of variants (Fig. 3B). Similarly, we did not observe differences between reclassification groups according to the time since clinical report (Fig. 4F). Therefore, we cannot establish an optimal timelapse to maximize the chances of getting a new diagnosis. This may reflect the limited sample size in the New Diagnosis group (4 new solves) limiting the ability to draw conclusions at this point, or it could reflect the serendipitous nature of new findings collected in public databases that do not follow any specific pattern. This point is illustrated by the AFG3L2 example, where two cases with very different timelines were solved by one new clinical genetics database entry (Fig. 5). Follow-up studies will help clarify any conclusions between length of Ann-Δ and likelihood of identifying a new causative finding.

The yield of new diagnoses via the RENEW approach is lower than previously reported manually reviewed data studies, but there are several key differences that may explain this discrepancy. RENEW was designed to facilitate systematic implementation of case review at scale, relying on availability of new data in ClinVar, HGMD, and OMIM. Thus, if new clinically relevant information has not been submitted or included in these resources (e.g. a publication not yet curated or in non-publicly available institutional databases), it will not be detected by our method. RENEW is not designed to replace an in-depth manual case review as described in previous reports but is optimized to identify clinical genetic findings that would have been easily identified if the relevant data were available at the time of the original clinical test.

Since the first use of the RENEW approach in our institution 2 years prior to the example described here, we have been able to continuously identify new diagnosis (Table 4) in a resource-efficient manner. Extrapolating from our example where an average of 5 variants were prioritized per case, with an average of 20 s per variant for review (100s per case), we can estimate the cost and benefit of this method at a larger scale. Considering a health system with 10,000 patients analyzed through RENEW, it will take 10⁶ s to review the resulting 50,000 variants every 6 months. This equates to approximately 272 total hours, or approximately one work week for a group of 7 experts. Based on the 2 year experience with RENEW at our institution, we expect ~ 0.9% newly identified causal findings, corresponding to 90 patients with newly resolved clinical genetics diagnoses in this example. Arguments can be made on either side as to whether dedicating just 7 variant curators for a week is cost effective for resolving 90 patients’ conditions every 6 months.

Table 4

Historical series of RENEW analysis indicating percentage of new diagnosis identified.
Date of RENEW reanalysis	Number of cases	New diagnosis	Percentage of new diagnosis	Total review time
Oct-20*	589	3	0.65%	25.14 h
Apr-21	461	7	1.52%	15.49 h
Oct-21	665	5	0.75%	20.65 h
Apr-22	1066	4	0.38%	32.06 h
*First use of RENEW. Additional changes to optimize the analysis were added in later iterations reflecting differences in total time per case.

In addition to identifying new diagnoses, it was recognized early in the development of RENEW that several of the prioritized variants were good candidates to explain the patient’s phenotype, but the evidence was still insufficient at that point to make a confident conclusion (e.g. variants in emerging gene-phenotype associations). While beyond of the scope of the RENEW process, these variants were collated as “Interesting Findings” and represent rich targets for subsequent in-depth study and may ultimately lead to additional solved cases. In this example, 63 (6.6%) cases fell into this category, and have been assigned to another study for extensive investigation and research profiling. In this sense, the RENEW approach could also be considered as a screening tool to identify candidate cases to allocate the always limited resources for research investigation.

In summary, we present here a novel approach to reannotate and efficiently analyze genomic data by focusing on novel information being introduced in gene and variant databases. RENEW combines automated annotation and prioritization steps with a rapid review process to efficiently identify relevant genetic findings that lead to new genetic diagnoses and variant reclassification. As clinical genetic testing increases in scale and scope, strategies such as RENEW will be critical to meet the challenges of data analysis and interpretation burden over time and provide patients with timely and highest quality of care.

Acknowledgements

We thank the patients’ families for allowing us to perform this research. This work was funded internally by the Center for Individualized Medicine at Mayo Clinic, Rochester MN.

Ethical Approval

Informed consent using a protocol approved by the Mayo Clinic institutional review board was obtained from all patients or patient representatives and can be provided upon request. All individual-level data has been appropriately deidentified. This research adheres to the Declaration of Helsinki.

Competing interests

Authors declare no conflict of interest regarding this work.

Authors' contributions

A.F., P.D. and E.W.K. conceived and lead the study. A.F., P.D., R.J.O., M.A.M. and E.W.K. wrote the article and developed the computational section of RENEW as described in the text. A.F., R.J.O., L.S.R., E.L.M., S.S. and J.A.M.R. performed the manual review step as described. M.A.C., C.K. and F.P.V. suggested additional improvements to RENEW. G.R.O., D.R. and M.W. provided additional computational support in the early developmental process. P.N.P., D.R.D., E.M., R.H.G., R.D., K.J.W., B.C.L., D.B.V., M.J.W. and L.A.S. identified the cases used to optimize our approach and the example cohort described in the manuscript. C.R.V., T.M.M., and K.N.L. contributed to the final version of the manuscript. All authors reviewed the manuscript and approved the submission.

Funding

This research was internally funded by the Center for Individualized Medicine at Mayo Clinic, Rochester (MN), U.S.A.

Availability of data and materials

All data, except patient’s personal health information, will be made available upon reasonable request and after completion of all legal and ethical procedures

AFA, S., H. R and G. P. (2015). "RepeatMasker Open-4.0." from http://www.repeatmasker.org.
Ashley, E. A. (2016). "Towards precision medicine." Nat Rev Genet17(9): 507-522.DOI: 10.1038/nrg.2016.86.
Asmann, Y. W., S. Middha, A. Hossain, S. Baheti, Y. Li, H. S. Chai, Z. Sun, P. H. Duffy, A. A. Hadad, A. Nair, X. Liu, Y. Zhang, E. W. Klee, K. R. Kalari and J. P. Kocher (2012). "TREAT: a bioinformatics tool for variant annotations and visualizations in targeted and exome sequencing data." Bioinformatics28(2): 277-278.DOI: 10.1093/bioinformatics/btr612.
Baker, S. W., J. R. Murrell, A. I. Nesbitt, K. B. Pechter, J. Balciuniene, X. Zhao, Z. Yu, E. H. Denenberg, E. T. DeChene, A. B. Wilkens, E. J. Bhoj, Q. Guan, M. C. Dulik, L. K. Conlin, A. N. Abou Tayoun, M. Luo, C. Wu, K. Cao, M. Sarmady, E. C. Bedoukian, J. Tarpinian, L. Medne, C. M. Skraban, M. A. Deardorff, I. D. Krantz, B. L. Krock and A. B. Santani (2019). "Automated Clinical Exome Reanalysis Reveals Novel Diagnoses." J Mol Diagn21(1): 38-48.DOI: 10.1016/j.jmoldx.2018.07.008.
Caporali, L., S. Magri, A. Legati, V. Del Dotto, F. Tagliavini, F. Balistreri, A. Nasca, C. La Morgia, M. Carbonelli, M. L. Valentino, E. Lamantea, S. Baratta, L. Schols, R. Schule, P. Barboni, M. L. Cascavilla, A. Maresca, M. Capristo, A. Ardissone, D. Pareyson, G. Cammarata, L. Melzi, M. Zeviani, L. Peverelli, C. Lamperti, S. B. Marzoli, M. Fang, M. Synofzik, D. Ghezzi, V. Carelli and F. Taroni (2020). "ATPase Domain AFG3L2 Mutations Alter OPA1 Processing and Cause Optic Neuropathy." Ann Neurol88(1): 18-32.DOI: 10.1002/ana.25723.
Epilepsy Genetics, I. (2019). "The Epilepsy Genetics Initiative: Systematic reanalysis of diagnostic exomes increases yield." Epilepsia60(5): 797-806.DOI: 10.1111/epi.14698.
Ewans, L. J., D. Schofield, R. Shrestha, Y. Zhu, V. Gayevskiy, K. Ying, C. Walsh, E. Lee, E. P. Kirk, A. Colley, C. Ellaway, A. Turner, D. Mowat, L. Worgan, M. L. Freckmann, M. Lipke, R. Sachdev, D. Miller, M. Field, M. E. Dinger, M. F. Buckley, M. J. Cowley and T. Roscioli (2018). "Whole-exome sequencing reanalysis at 12 months boosts diagnosis and is cost-effective when applied early in Mendelian disorders." Genet Med20(12): 1564-1574.DOI: 10.1038/gim.2018.39.
Farwell, K. D., L. Shahmirzadi, D. El-Khechen, Z. Powis, E. C. Chao, B. Tippin Davis, R. M. Baxter, W. Zeng, C. Mroske, M. C. Parra, S. K. Gandomi, I. Lu, X. Li, H. Lu, H. M. Lu, D. Salvador, D. Ruble, M. Lao, S. Fischbach, J. Wen, S. Lee, A. Elliott, C. L. Dunlop and S. Tang (2015). "Enhanced utility of family-centered diagnostic exome sequencing with inheritance model-based analysis: results from 500 unselected families with undiagnosed genetic conditions." Genet Med17(7): 578-586.DOI: 10.1038/gim.2014.154.
Ji, J., M. L. Leung, S. Baker, J. L. Deignan and A. Santani (2021). "Clinical Exome Reanalysis: Current Practice and Beyond." Mol Diagn Ther25(5): 529-536.DOI: 10.1007/s40291-021-00541-7.
Karczewski, K. J., L. C. Francioli, G. Tiao, B. B. Cummings, J. Alfoldi, Q. Wang, R. L. Collins, K. M. Laricchia, A. Ganna, D. P. Birnbaum, L. D. Gauthier, H. Brand, M. Solomonson, N. A. Watts, D. Rhodes, M. Singer-Berk, E. M. England, E. G. Seaby, J. A. Kosmicki, R. K. Walters, K. Tashman, Y. Farjoun, E. Banks, T. Poterba, A. Wang, C. Seed, N. Whiffin, J. X. Chong, K. E. Samocha, E. Pierce-Hoffman, Z. Zappala, A. H. O'Donnell-Luria, E. V. Minikel, B. Weisburd, M. Lek, J. S. Ware, C. Vittal, I. M. Armean, L. Bergelson, K. Cibulskis, K. M. Connolly, M. Covarrubias, S. Donnelly, S. Ferriera, S. Gabriel, J. Gentry, N. Gupta, T. Jeandet, D. Kaplan, C. Llanwarne, R. Munshi, S. Novod, N. Petrillo, D. Roazen, V. Ruano-Rubio, A. Saltzman, M. Schleicher, J. Soto, K. Tibbetts, C. Tolonen, G. Wade, M. E. Talkowski, C. Genome Aggregation Database, B. M. Neale, M. J. Daly and D. G. MacArthur (2020). "The mutational constraint spectrum quantified from variation in 141,456 humans." Nature581(7809): 434-443.DOI: 10.1038/s41586-020-2308-7.
Klee, E. W., M. A. Cousin, E. V. F. Pinto, J. A. Morales-Rosado, E. L. Macke, W. G. Jenkinson, A. Ferrer, L. E. Schultz-Rogers, R. J. Olson, G. R. Oliver, A. N. Sigafoos, T. L. Schwab, M. T. Zimmermann, R. A. Urrutia, C. Kaiwar, A. Gupta, P. R. Blackburn, N. J. Boczek, C. A. Prochnow, R. J. Lowy, L. A. Mulvihill, T. M. McAllister, S. L. Aoudia, T. M. Kruisselbrink, L. B. Gunderson, J. L. Kemppainen, L. J. Fisher, J. M. Tarnowski, M. M. Hager, S. A. Kroc, N. L. Bertsch, K. E. Agre, J. L. Jackson, S. K. Macklin-Mantia, M. I. Murphree, L. M. Rust, J. M. Summer Bolster, S. A. Beck, P. S. Atwal, M. S. Ellingson, S. S. Barnett, K. J. Rasmussen, C. A. Lahner, Z. Niu, L. Hasadsri, M. J. Ferber, C. A. Marcou, K. J. Clark, P. N. Pichurin, D. R. Deyle, E. Morava-Kozicz, R. H. Gavrilova, R. Dhamija, K. J. Wierenga, B. C. Lanpher, D. Babovic-Vuksanovic, G. Farrugia, L. A. Schimmenti, A. K. Stewart and K. N. Lazaridis (2021). "Impact of integrated translational research on clinical exome sequencing." Genet Med23(3): 498-507.DOI: 10.1038/s41436-020-01005-9.
Kocher, J. P., D. J. Quest, P. Duffy, M. A. Meiners, R. M. Moore, D. Rider, A. Hossain, S. N. Hart and V. Dinu (2014). "The Biological Reference Repository (BioR): a rapid and flexible system for genomics annotation." Bioinformatics30(13): 1920-1922.DOI: 10.1093/bioinformatics/btu137.
Landrum, M. J., J. M. Lee, M. Benson, G. R. Brown, C. Chao, S. Chitipiralla, B. Gu, J. Hart, D. Hoffman, W. Jang, K. Karapetyan, K. Katz, C. Liu, Z. Maddipatla, A. Malheiro, K. McDaniel, M. Ovetsky, G. Riley, G. Zhou, J. B. Holmes, B. L. Kattman and D. R. Maglott (2018). "ClinVar: improving access to variant interpretations and supporting evidence." Nucleic Acids Res46(D1): D1062-D1067.DOI: 10.1093/nar/gkx1153.
Lee, H., J. L. Deignan, N. Dorrani, S. P. Strom, S. Kantarci, F. Quintero-Rivera, K. Das, T. Toy, B. Harry, M. Yourshaw, M. Fox, B. L. Fogel, J. A. Martinez-Agosto, D. A. Wong, V. Y. Chang, P. B. Shieh, C. G. Palmer, K. M. Dipple, W. W. Grody, E. Vilain and S. F. Nelson (2014). "Clinical exome sequencing for genetic identification of rare Mendelian disorders." JAMA312(18): 1880-1887.DOI: 10.1001/jama.2014.14604.
Liu, P., L. Meng, E. A. Normand, F. Xia, X. Song, A. Ghazi, J. Rosenfeld, P. L. Magoulas, A. Braxton, P. Ward, H. Dai, B. Yuan, W. Bi, R. Xiao, X. Wang, T. Chiang, F. Vetrini, W. He, H. Cheng, J. Dong, C. Gijavanekar, P. J. Benke, J. A. Bernstein, T. Eble, Y. Eroglu, D. Erwin, L. Escobar, J. B. Gibson, K. Gripp, S. Kleppe, M. K. Koenig, A. M. Lewis, M. Natowicz, P. Mancias, L. Minor, F. Scaglia, C. P. Schaaf, H. Streff, H. Vernon, C. L. Uhles, E. H. Zackai, N. Wu, V. R. Sutton, A. L. Beaudet, D. Muzny, R. A. Gibbs, J. E. Posey, S. Lalani, C. Shaw, C. M. Eng, J. R. Lupski and Y. Yang (2019). "Reanalysis of Clinical Exome Sequencing Data." N Engl J Med380(25): 2478-2480.DOI: 10.1056/NEJMc1812033.
Masuda, T., X. Wang, M. Maeda, M. C. Canver, F. Sher, A. P. Funnell, C. Fisher, M. Suciu, G. E. Martyn, L. J. Norton, C. Zhu, R. Kurita, Y. Nakamura, J. Xu, D. R. Higgs, M. Crossley, D. E. Bauer, S. H. Orkin, P. V. Kharchenko and T. Maeda (2016). "Transcription factors LRF and BCL11A independently repress expression of fetal hemoglobin." Science351(6270): 285-289.DOI: 10.1126/science.aad3312.
Munz, M., E. Ruark, A. Renwick, E. Ramsay, M. Clarke, S. Mahamdallie, V. Cloke, S. Seal, A. Strydom, G. Lunter and N. Rahman (2015). "CSN and CAVA: variant annotation tools for rapid, robust next-generation sequencing analysis in the clinical setting." Genome Med7: 76.DOI: 10.1186/s13073-015-0195-6.
Nambot, S., J. Thevenon, P. Kuentz, Y. Duffourd, E. Tisserant, A. L. Bruel, A. L. Mosca-Boidron, A. Masurel-Paulet, D. Lehalle, N. Jean-Marcais, M. Lefebvre, P. Vabres, S. El Chehadeh-Djebbar, C. Philippe, F. Tran Mau-Them, J. St-Onge, T. Jouan, M. Chevarin, C. Poe, V. Carmignac, A. Vitobello, P. Callier, J. B. Riviere, L. Faivre, C. Thauvin-Robinet and G. Orphanomix Physicians (2018). "Clinical whole-exome sequencing for the diagnosis of rare disorders with congenital anomalies and/or intellectual disability: substantial interest of prospective annual reanalysis." Genet Med20(6): 645-654.DOI: 10.1038/gim.2017.162.
O'Daniel, J. M., H. M. McLaughlin, L. M. Amendola, S. J. Bale, J. S. Berg, D. Bick, K. M. Bowling, E. C. Chao, W. K. Chung, L. K. Conlin, G. M. Cooper, S. Das, J. L. Deignan, M. O. Dorschner, J. P. Evans, A. A. Ghazani, K. A. Goddard, M. Gornick, K. D. Farwell Hagman, T. Hambuch, M. Hegde, L. A. Hindorff, I. A. Holm, G. P. Jarvik, A. Knight Johnson, L. Mighion, M. Morra, S. E. Plon, S. Punj, C. S. Richards, A. Santani, B. H. Shirts, N. B. Spinner, S. Tang, K. E. Weck, S. M. Wolf, Y. Yang and H. L. Rehm (2017). "A survey of current practices for genomic sequencing test interpretation and reporting processes in US laboratories." Genet Med19(5): 575-582.DOI: 10.1038/gim.2016.152.
Ohishi, A., Y. Masunaga, S. Iijima, K. Yamoto, F. Kato, M. Fukami, H. Saitsu and T. Ogata (2020). "De novo ZBTB7A variant in a patient with macrocephaly, intellectual disability, and sleep apnea: implications for the phenotypic development in 19p13.3 microdeletions." J Hum Genet65(2): 181-186.DOI: 10.1038/s10038-019-0690-5.
Posey, J. E., T. Harel, P. Liu, J. A. Rosenfeld, R. A. James, Z. H. Coban Akdemir, M. Walkiewicz, W. Bi, R. Xiao, Y. Ding, F. Xia, A. L. Beaudet, D. M. Muzny, R. A. Gibbs, E. Boerwinkle, C. M. Eng, V. R. Sutton, C. A. Shaw, S. E. Plon, Y. Yang and J. R. Lupski (2017). "Resolution of Disease Phenotypes Resulting from Multilocus Genomic Variation." N Engl J Med376(1): 21-31.DOI: 10.1056/NEJMoa1516767.
Ramoni, R. B., J. J. Mulvihill, D. R. Adams, P. Allard, E. A. Ashley, J. A. Bernstein, W. A. Gahl, R. Hamid, J. Loscalzo, A. T. McCray, V. Shashi, C. J. Tifft, N. Undiagnosed Diseases and A. L. Wise (2017). "The Undiagnosed Diseases Network: Accelerating Discovery about Health and Disease." Am J Hum Genet100(2): 185-192.DOI: 10.1016/j.ajhg.2017.01.006.
Richards, S., N. Aziz, S. Bale, D. Bick, S. Das, J. Gastier-Foster, W. W. Grody, M. Hegde, E. Lyon, E. Spector, K. Voelkerding, H. L. Rehm and A. L. Q. A. Committee (2015). "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 Med17(5): 405-424.DOI: 10.1038/gim.2015.30.
Salfati, E. L., E. G. Spencer, S. E. Topol, E. D. Muse, M. Rueda, J. R. Lucas, G. N. Wagner, S. Campman, E. J. Topol and A. Torkamani (2019). "Re-analysis of whole-exome sequencing data uncovers novel diagnostic variants and improves molecular diagnostic yields for sudden death and idiopathic diseases." Genome Med11(1): 83.DOI: 10.1186/s13073-019-0702-2.
Seo, G. H., H. Lee, J. Lee, H. Han, Y. K. Cho, M. Kim, Y. Choi, J. Choi, I. H. Choi, S. Rhie, K. Y. Chae, Y. M. Kim, C. K. Cheon, S. J. Kim, J. Lee, E. Kang, J. H. Byeon, H. J. Yu, Y. L. Shin, A. Oh, W. J. Kim, M. S. Yum, B. H. Lee and B. L. Eun (2022). "Diagnostic performance of automated, streamlined, daily updated exome analysis in patients with neurodevelopmental delay." Mol Med28(1): 38.DOI: 10.1186/s10020-022-00464-x.
Stenson, P. D., E. V. Ball, M. Mort, A. D. Phillips, J. A. Shiel, N. S. Thomas, S. Abeysinghe, M. Krawczak and D. N. Cooper (2003). "Human Gene Mutation Database (HGMD): 2003 update." Hum Mutat21(6): 577-581.DOI: 10.1002/humu.10212.
von der Lippe, C., K. Tveten, T. E. Prescott, O. L. Holla, O. L. Busk, K. B. Burke, F. H. Sansbury, J. Baptista, A. E. Fry, D. Lim, S. Jolles, J. Evans, D. Osio, C. Macmillan, I. Bruno, F. Faletra, S. Climent, R. Urreitzi, J. Hoenicka, F. Palau, A. S. A. Cohen, K. Engleman, D. Zhou, S. M. Amudhavalli, M. Jeanne, F. Bonnet-Brilhault, J. Levy, S. Drunat, N. Derive, M. G. Haug and W. M. Thorstensen (2022). "Heterozygous variants in ZBTB7A cause a neurodevelopmental disorder associated with symptomatic overgrowth of pharyngeal lymphoid tissue, macrocephaly, and elevated fetal hemoglobin." Am J Med Genet A188(1): 272-282.DOI: 10.1002/ajmg.a.62492.
Wenger, A. M., H. Guturu, J. A. Bernstein and G. Bejerano (2017). "Systematic reanalysis of clinical exome data yields additional diagnoses: implications for providers." Genet Med19(2): 209-214.DOI: 10.1038/gim.2016.88.
Yang, Y., D. M. Muzny, F. Xia, Z. Niu, R. Person, Y. Ding, P. Ward, A. Braxton, M. Wang, C. Buhay, N. Veeraraghavan, A. Hawes, T. Chiang, M. Leduc, J. Beuten, J. Zhang, W. He, J. Scull, A. Willis, M. Landsverk, W. J. Craigen, M. R. Bekheirnia, A. Stray-Pedersen, P. Liu, S. Wen, W. Alcaraz, H. Cui, M. Walkiewicz, J. Reid, M. Bainbridge, A. Patel, E. Boerwinkle, A. L. Beaudet, J. R. Lupski, S. E. Plon, R. A. Gibbs and C. M. Eng (2014). "Molecular findings among patients referred for clinical whole-exome sequencing." JAMA312(18): 1870-1879.DOI: 10.1001/jama.2014.14601.

No competing interests reported.

SuppFig1.pdf

Download PDF

Journal Publication

published 27 Mar, 2024

Read the published version in Human Genetics →

Editorial decision: Revision requested
14 Dec, 2023
Reviews received at journal
04 Dec, 2023
Reviewers agreed at journal
07 Nov, 2023
Reviewers agreed at journal
10 Aug, 2023
Reviews received at journal
26 Jul, 2023
Reviewers agreed at journal
26 Jul, 2023
Reviewers invited by journal
19 Jul, 2023
Editor assigned by journal
11 Jul, 2023
Submission checks completed at journal
11 Jul, 2023
First submitted to journal
06 Jul, 2023

You are reading this latest preprint version

Semiautomated approach focused on new genomic information results in time and effort- efficient reannotation of negative exome data

Status:

Journal Publication

Version 1

Abstract

Introduction:

Methods

Results

Discussion

Figures

INTRODUCTION

MATERIAL AND METHODS

Patient sequencing data

Reannotation of sequencing data

Variant Filtering and Prioritization Methods

Manual review of prioritized variants

RESULTS

Cohort

Validation of RENEW approach with independently solved cases

Variant filtering and prioritization

Manual review of prioritized variants

RENEW process examples:

Example: Emerging AFG3L2 disease association

DISCUSSION

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1