somatic slc35a2 mosaicism correlates with clinical ﬁ

ARTICLE OPEN ACCESS

Somatic SLC35A2 mosaicism correlates withclinical findings in epilepsy brain tissueKatherine E Miller PhD Daniel C Koboldt MS Kathleen M Schieffer PhD Tracy A Bedrosian PhD

Erin Crist MMSc LGC Adrienne Sheline MPH Kristen Leraas MS Vincent Magrini PhD Huachun Zhong MS

Patrick Brennan MS Jocelyn Bush MS James Fitch BS Natalie Bir BS Anthony R Miller PhD

Catherine E Cottrell PhD Jeffrey Leonard MD Jonathan A Pindrik MD Jerome A Rusin MD

Summit H Shah MD MPH Peter White PhD Richard K Wilson PhD Elaine R Mardis PhD

Christopher R Pierson MD PhD and Adam P Ostendorf MD

Neurol Genet 20206e460 doi101212NXG0000000000000460

Correspondence

Dr Ostendorf

AdamOstendorf

nationwidechildrensorg

AbstractObjectiveMany genetic studies of intractable epilepsy in pediatric patients primarily focus on inheritedconstitutional genetic deficiencies identified in patient blood Recently studies have revealedsomatic mosaicism associated with epilepsy in which genetic variants are present only ina subset of brain cells We hypothesize that tissue-specific somatic mosaicism represents animportant genetic etiology in epilepsy and aim to discover somatic alterations in epilepsy-affected brain tissue

MethodsWe have pursued a research study to identify brain somatic mosaicism using next-generationsequencing (NGS) technologies in patients with treatment refractory epilepsy who haveundergone surgical resection of affected brain tissue

ResultsWe used an integrated combination of NGS techniques and conventional approaches (radi-ology histopathology and electrophysiology) to comprehensively characterize multiple brainregions from a single patient with intractable epilepsy We present a 3-year-old male patientwith West syndrome and intractable tonic seizures in whom we identified a pathogenicframeshift somatic variant in SLC35A2 present at a range of variant allele fractions(42ndash195) in 12 different brain tissues detected by targeted sequencing The proportion ofthe SLC35A2 variant correlated with severity and location of neurophysiology and neuro-imaging abnormalities for each tissue

ConclusionsOur findings support the importance of tissue-based sequencing and highlight a correlation inour patient between SLC35A2 variant allele fractions and the severity of epileptogenic phe-notypes in different brain tissues obtained from a grid-based resection of clinically definedepileptogenic regions

From the The Steve and Cindy Rasmussen Institute for Genomic Medicine (KEM DCK KMS TAB EC KL VM HZ PB JB JF NB ARM CEC PW RKW ERM)Abigail Wexner Research Institute at Nationwide Childrenrsquos Hospital Columbus OH Division of Genetic and Genomic Medicine (EC) Nationwide Childrenrsquos Hospital Columbus OHDepartment of Neurosurgery (AS JL JAP) Nationwide Childrenrsquos Hospital Columbus OH Department of Pathology and Laboratory Medicine (CRP) Nationwide ChildrenrsquosHospital Columbus OH Division of Child Neurology (APO) Nationwide Childrenrsquos Hospital Columbus OH Department of Radiology (JAR SHS) Nationwide Childrenrsquos HospitalColumbus OH Department of Pediatrics (DCK VM CEC JL PW RKW ERM APO) The Ohio State University College of Medicine Columbus OH Department ofNeurosurgery (JL JAP APO) The Ohio State University College of Medicine Columbus OH Department of Pathology (CEC CRP) The Ohio State University College of MedicineColumbus OH and Department of Biomedical Education amp Anatomy (CRP) Division of Anatomy The Ohio State University College of Medicine Columbus OH

Go to NeurologyorgNG for full disclosures Funding information is provided at the end of the article

The Article Processing Charge was funded by the authors

This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 40 (CC BY-NC-ND) which permits downloadingand sharing the work provided it is properly cited The work cannot be changed in any way or used commercially without permission from the journal

Copyright copy 2020 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the American Academy of Neurology 1

Many genetic epilepsies are characterized by Mendelian in-heritance notably in ion channel genes (eg SCN12AKCNQ2 and GRIN12A2B2D) which also may occur denovo in individuals without any family history of epilepticdisorders1ndash3 Approximately one-third of epilepsies aretreatment refractory due to failure of medication to alleviateor ameliorate seizures4ndash6 In select cases electrophysiologicbrain mapping can identify affected regions and the re-section or disconnection of these regions is potentially cu-rative For some patients the likelihood of becomingseizure-free after surgical resection or disconnection is ashigh as 8056

Several studies have recently identified somatic mosaicismassociated with epilepsy in which genetic variants are presentonly in a subset of brain cells identified in resected tissue Themost robust association involves genes in the PI3K-AKT-mTOR pathway wherein hyperactivation of the mTORpathway occurs in patients with hemimegalencephaly andfocal cortical dysplasia (FCD)7ndash10 Still the genetic etiology ofmany epileptic phenotypes associated with somatic mosai-cism remains ill defined

At Nationwide Childrenrsquos Hospital a National Associationof Epilepsy Centers Level 4 epilepsy center we have estab-lished a research program to identify brain somatic mosai-cism in patients with treatment refractory epilepsy who haveundergone surgical resection (figure e-1 linkslwwcomNXGA273) High-depth next-generation sequencing(NGS) is performed on brain DNA in comparison to blood-derived DNA Somatic variants are assessed for pathoge-nicity and analyzed in the context of clinically obtainedmultimodal information We hypothesize that tissue-specific somatic mosaicism represents an important ge-netic etiology in epilepsy

MethodsStandard protocol approvals registrationsand patient consentsThis study was designed to perform genomic profiling ofexcised tissue and matched (blood) samples from a patientwith refractory epilepsy Written informed consent wasobtained and the patient was enrolled as part of aninstitutional review board-approved research study (IRB18-00786) within The Steve and Cindy Rasmussen Institute forGenomic Medicine at Nationwide Childrenrsquos Hospital

Data availabilityThe SLC35A2 variant information was deposited inClinVar (ncbinlmnihgovclinvar) under accessionnumber VCV0006351171

HistopathologyHematoxylin and eosin staining was performed as part ofroutine clinical workup according to recommendations of theInternational League Against Epilepsy11

NeurophysiologyThe index patient underwent multiple extracranial EEG anda single intracranial EEG All data were digitally acquired andinterpreted with a Nihon Kohden system Interpretationswere performed by a board-certified clinical neurophysiologistaccording to standard clinical care Quantification of epilep-tiform activity was performed by counting spikes polyspikesand ripples during a 5-minute sample at the beginning of eachhour during the last day of recording (120 total minutes from24 epochs) For each tissue the correlation between activity(average spikes per minute) and variant allele fraction (VAF)(average tissue VAF from NEBNext Direct [NND]) wascalculated in R using the cortest operation

MRIThe index patient was imaged 3 times before the final diagnosisThe first 2 scans were obtained on a GE Discovery MR750 3Tmagnet (GE Medical Systems Milwaukee WI) using a 32-channel head coil with sedation performed by a pediatric anes-thesiologist The first 2 scans included 3 plane brain volumeimaging long inversion recovery (T1 volumetric) axialcoronalT2 FSE axial susceptibility weighted and axial diffusionsequences The second scan also included 3 plane T2 fluid-attenuated inversion recovery (FLAIR) CUBE (T2 volumetric)axial diffusion tensor imaging using 30 gradient encodingdirections and postgadobutrol (Gadovist Bayer HealthCareWhippany NJ) coronal T2 FLAIR 3 plane brain volume im-aging long inversion recovery (T1 volumetric) and axial T1 spinecho sequences The third MRI scan was performed on a Sie-mens PRISMA 3Tmagnet (Siemens Erlangen Germany) usinga 32-channel head coil with sedation by a pediatric anesthesi-ologist The third scan included the following 3 plane magne-tization prepared-rapid gradient ECHO (T1 volumetric) 3plane T2 FLAIR sampling perfection with application-optimized(T2 volumetric) axial diffusion tensor imaging using 30 gradientencoding directions axial and coronal T2 high resolution axialsusceptibility-weighted imaging and postgadobutrol (GadovistBayer HealthCare Whippany NJ) coronal T2 FLAIR 3 plane

GlossaryACMG = American College of Medical Genetics and Genomics AMP = Association for Molecular Pathology FCD = focalcortical dysplasia FDG = fluorodeoxyglucose FFPE = formalin-fixed paraffin embedded FPKM = Fragments Per Kilobase oftranscript per Million IGV = Integrated Genomics Viewer NGS = next-generation sequencing NMD = nonsense-mediateddecay NND = NEBNext Direct UMI = unique molecular identifier VAF = variant allele fraction VUS = variant of unknownsignificance

2 Neurology Genetics | Volume 6 Number 4 | August 2020 NeurologyorgNG

magnetization prepared-rapid gradient ECHO (T1 volumetric)and axial T1 spin echo sequences Interpretations were per-formed by a board-certified radiologist

Fluorodeoxyglucose-PET scanAfter IV injection of fluorodeoxyglucose (FDG) radiotracera PET of the head was performed using a 64-slice GEHealthcare (Chicago IL) Discovery-690 scanner with the pa-tient under sedation and with simultaneous EEG monitoringThe images were postprocessed and fused with the MRI of thehead from the same day usingMIM (Cleveland OH) software

Surgical resectionThe patient underwent a 2-stage epilepsy surgery wherebystage 1 consisted of insertion of a subdural convexity grid andstrip electrodes to identify epileptogenic zones and stage 2consisted of a grid-based resection of epileptogenic zones in-cluding the left temporal lobe left amygdala and hippocampusleft inferior parietal lobe left occipital lobe left medial occipitallobe left inferior parietal cortex and left superior temporal gy-rus Twelve tissues from different resected brain sections froma singular patient were used for co-DNARNA extraction forthe purpose of study by NGS (table e-1 linkslwwcomNXGA273) For frozen samples we used 30 mg of each tissue forDNA extraction For formalin-fixed paraffin-embedded (FFPE)samples we used a calculation to determine the number ofscrolls needed based on the total amount of tissue in the blockso the amount of DNA extracted was comparable to all othertissue blocks DNA extracted from the patientrsquos peripheral bloodwas used for inherited (germline) variant calling and as a com-parator for analysis of acquired (somatic) variants IntracranialEEG was performed using Ad-Tech (Oak Creek WI) subduralgrid and strip electrodes implanted during a left-sided craniot-omy Electrophysiologic data were digitally recorded usinga Nihon Kohden (Tokyo Japan) digital EEG recording system

Exome and transcriptome sequencingDNA extracted from peripheral blood and DNARNA frombrain resection tissueswas used for enhanced exome sequencingusing the NEBNext Ultra II FS DNA Library Prep Kit (NewEngland BioLabs Ipswich MA) IDT xGen Exome ResearchPanel v10 enhanced with the xGenCNV Backbone Panel-TechAccess (Integrated DNATechnologies Coralville IA) was usedfor target enrichment by hybrid capture Paired-end 151-bpreads were generated on Illumina HiSeq 4000 Alignment tohuman reference genome build GRCh37 and secondary anal-ysis was performed using a custom in-house pipeline12 Germ-line variants were called using GATKrsquos HaplotypeCaller andsomatic variants were called using MuTect21314 Germlinevariants were filtered based on the following characteristicsGATKquality score (ge30) population frequency (le 1) depthof sequencing (ge8 reads) VAF (ge 20) protein-coding regionor distance from canonical splice site le3 bp damaging pre-diction scores (Combined Annotation Dependent Depletionge15 Sorting Intolerant From Tolerant le005 Genomic Evo-luationary Rate Profiling++ ge 2) and relevance of gene-diseaseassociation with patient phenotype using Online Mendelian

Inheritance inMan (OMIM) and ClinVar databases Candidategermline variants were assessed for pathogenicity according tothe American College of Medical Genetics and Genomics(ACMG) and Association for Molecular Pathology (AMP)variant interpretation guidelines15

Somatic variants were filtered based on the following charac-teristics MuTect2 = PASS GATK quality score (ge30)protein-coding region or distance from canonical splice site le3bp depth of sequencing (ge8 total reads) absence from bloodcomparator alternate allele reads (ge5) and VAF (ge1) Var-iants passing all criteria were then manually reviewed in In-tegrated Genomics Viewer (IGV) to assess for strand biasread-end bias or other biases which might suggest sequencingartifacts Candidate somatic variants were assessed for patho-genicity and for association with epilepsy or a seizure-relatedphenotype using OMIM and ClinVar databases

Brain-extracted RNA was subject to DNase treatment andribodepletion before library construction using IlluminarsquosTruSeq Stranded total RNA kit Paired-end 151-bp reads weregenerated on Illumina HiSeq 4000 and aligned to human ref-erence genome build GRCh38 Alignment and variant callingwas performed using a custom in-house pipeline and the splice-aware aligner STAR10 Cufflinks (version 221) was used togenerate gene-based Fragments Per Kilobase of transcript perMillion mapped reads (FPKM) values for SLC35A2 For figuree-2 linkslwwcomNXGA273 other epilepsy brain tissueswere previously resected from patients enrolled on our in-stitutional review board protocol DESeq2 was used for nor-malization of expression values and values for SLC35A2 wereplotted for 6 tissues (A1 A2 B C D1 and D2) from thedescribed patient vs brain tissues from other epilepsy patients16

Targeted sequencingGenomic DNA (10 ng) was used as a template for PCRSLC35A2 primers forward (59-AGCCTGAGCTGCCTTTG-39) and reverse (59- CGGCCACTGGATCAGAAC-39) PCRwas performed using 2x Q5 MM (New England BioLabs) and200 nM primers with the following conditions 300 at 98degC 30cycles of 100 at 98degC 200 at 58degC 200 at 72degC and a finalextension of 59 at 72degC Amplified products were purified using18X SPRIselect followed by end-repair and dA-tailing usingNEBNext Ultra II DNA Library Prep kit reagents The reactionwas followed by adapter ligation with unique molecular iden-tifier (UMI)-IDT-indexed adaptors (Integrated DNA Tech-nologies) Adaptor-ligated samples were purified using 12XSPRIselect and used for library amplification with Q5MM andIllumina P5P7 primer mix A post-PCR 12X SPRIselectcleanup was performed and libraries were pooled and se-quenced on Illumina HiSeq4000

NND is a hybridization-based target-enrichment approachfor direct sequencing of region(s) of interest17 We useda custom-designed kit targeting coding regions of SLC35A2for deep sequencing of DNA and RNA We used 100 ng(frozen tissue DNA and blood DNA) 200 ng (frozen tissue

NeurologyorgNG Neurology Genetics | Volume 6 Number 4 | August 2020 3

RNA) or 500 ng (FFPE tissue DNA) For RNA librariesdouble-stranded cDNA was synthesized using high-capacitycDNA reverse transcript kit for first strand (Thermo FisherScientific and Applied Biosystems) followed by NEBNextUltra II nondirectional RNA second strand synthesis kit (NewEngland BioLabs Inc) The protocol was then followedaccording to the manufacturerrsquos suggestions which in-corporated unique dual indexes with a UMI next to the i5index DNA libraries were sequenced on Illumina iSeqwhereas cDNA libraries were sequenced on IlluminaMiniSeqNND demo pipeline was used for data processing (githubcomDirectedGenomicsDemoPipeline) The raw dedupli-cated BAM files were used with SAMtools mpileup andmpileup2cns VarScan (version 234) commands to generateread counts for our variant of interest1819

ResultsDisease presentation and clinical findingsThe patient is a 3-year-old male who presented at age 9months with West syndrome a severe form of epilepsycharacterized by infantile spasms an abnormal EEG patterntermed hypsarrhythmia and developmental delay The eti-ologies for West syndrome are multifactorial and may include

genetic or structural abnormalities20 A commercially availableepilepsy-specific gene panel using patient blood did notidentify any pathogenic variants and the patientrsquos initialstructural brain MRI was normal His EEG revealed hypsar-rhythmia with a left posterior quadrant predominance ofepileptiform activity and generalized epileptic spasms He wastreated with adrenocorticotropin hormone with initial re-mission followed 5 months later by recurrence and remissionwith prednisolone therapy At age 25 years epileptic spasmsrecurred along with focal tonic seizures which were refractoryto several different antiseizure medications At that time heunderwent long-term EEG monitoring FDG-PET MRI andneurodevelopmental testing as part of an evaluation for re-fractory seizures New findings included seizures with leftposterior quadrant onset seizures a brain MRI finding sug-gestive of a temporal lobe dysplasia and FDG-PET hypo-metabolism in the left temporoparietal region This wasfollowed by placement of subdural grids and strips coveringthe left posterior frontal temporal parietal and occipitallobes Ictal findings defined as spike and wave and gammafrequency activity were diffuse in the electrodes over the lefttemporal and parieto-occiptal regions including the mesialtemporal contacts (figure 1) A gradient was present in theinterictal epileptiform activity where spikes were nearly

Figure 1 Epileptic spasm and intracranial grid location

(A) Ictal recording of typical epileptic spasms Thespikes and high-frequency activity at ictal onsetwere broadly distributed in the mesial temporalfrontotemporal inferior parietal and occipitalcontacts (sensitivity 75 uV low-frequency filter53 Hz high-frequency filter 500 Hz) (B) Place-ment of the grid and strips (blue) over the leftfrontal temporal parietal and occipital lobes (C)Predominant ictal onset (red) and rapid propa-gation (orange) over a broad region (D) Locationof the tissue samples and spike quantification FT= frontotemporal MT = mesial temporal Occ =occipital


continuous or abundant in the contacts over the hippocampusand temporal lobe contacts and frequent or occasional incontacts over the parietal and portions of the occipital lobeBecause of the diffuse nature of the ictal pattern involving theleft posterior quadrant the patient underwent a grid-basedresection of presumed epileptic zones including removal ofthe left temporal lobe left amygdala left hippocampus leftinferior parietal lobe and left occipital lobe (table e-1 linkslwwcomNXGA273) Histologic examination of the resec-ted tissue specimens showed cortical dyslamination with areasfeaturing neurons arranged in microcolumns (occipital lobe)and areas with ill-defined hexalaminar architecture (temporaland occipital lobes) consistent with FCD type 1c accordingto the International League Against Epilepsy classificationsystem11 The amygdala and hippocampus showed fragmentsof gray and white matter with gliosis In summary figure 2shows neuroimaging EEG and histopathology of 2 vastlydifferentially affected brain regions the hippocampus andoccipital lobe

Exome sequencing of brain tissues revealssomatic mosaicism in SLC35A2DNA was extracted from 6 brain tissues from differentregions within the same patient brain and from peripheralblood mononuclear cells as a matched comparator All

brain tissues were from suspected affected areas andresected during the same surgery We performed exomesequencing initially on the 6 brain tissues plus matchedblood to search for inherited (germline) variation and ac-quired (somatic) variation in coding regions includingsingle nucleotide insertiondeletion and copy numbervariants For the temporal lobe and occipital lobe we se-quenced 2 different samples removed from the same brainregion Mean on-target exome sequencing read coveragedepth was 222x for matched blood and 331x (range282xndash384x) for brain tissues (table e-2 linkslwwcomNXGA273) Our analysis of inherited variants yielded 3variants that merited further review (table e-3 linkslwwcomNXGA273) Two variants were within genes un-related to the patientrsquos phenotype The third variant was inDYNC1H1 and was previously identified in our patient aswell as the patientrsquos unaffected mother through clinicalgenetic testing and reported as a variant of unknown sig-nificance (VUS) De novo variants in DYNC1H1 have beenidentified in individuals with malformations of corticaldevelopment and in individuals with epileptic encephalo-pathies including West syndrome2122 Because of theinherited status of this variant we do not believe that it isassociated with the patientrsquos phenotype and assessed it asa VUS

Figure 2 Neuroimaging EEG and histopathology of differentially affected hippocampus and parietaloccipital lobe

(A) Hippocampus T2-weighted coronal MRI of the brain demonstrates increased signal within the left temporal white matter with circling of the asymmetricgray-white matter blurring on the left temporal lobe (top left) FDG-PET fused with MRI demonstrates focal area of asymmetrically decreased uptake withinthe left temporal lobe and hippocampus area arrows point to the left temporal lobe and left hippocampal region (top right) Intracranial EEG demonstrateshigh-amplitude spikes (asterisks) abundantly in the hippocampus contacts (bottom left) Histology image at 100times HampE-stained section from hippocampusshows expected cytoarchitecture (bottom right) (B) Parietaloccipital lobe T1-weighted coronalMRI of the brain demonstrates no focal abnormality in the leftparietal lobe (top left) FDG-PET fused with MRI demonstrates focal area of asymmetrically decreased uptake within the left parietal lobe arrows point to theleft parietal lobe (top right) Intracranial EEG demonstrates high-amplitude spikes (asterisks) only occasionally in the occipital contacts (bottom left) Histologyimage at 40times HampE-stained section shows cortical dyslamination with neurons of the occipital neocortex arranged in microcolumns (arrowheads) FDG =fluorodeoxyglucose HampE = hematoxylin and eosin


Our analysis of acquired somatic variants revealed 46 variantsthat met the initial filtering criteria but after manuallyreviewing these in the IGV to discount strand bias or othersequencing artifacts we determined that 4146 suspect var-iants were sequencing artifacts The 5 remaining candidatesomatic mosaic variants are listed in table e-4 linkslwwcomNXGA273 We identified a likely causal variant in the solutecarrier family 35 member A2 (SLC35A2) gene which wasabsent in the matched blood comparator and present ina range of VAFs from 06 to 277 in the resected braintissues (table 1) The X-linked NM_0056603c634_635delTC (pSer212LeufsTer9) variant is predicted to causea frameshift in the transmembrane domain and a prematurestop codon 9 amino acid downstream This variant has notbeen reported in the Genome Aggregation Database of150647 alleles from the general population23 We found 1report of this somatic variant in a patient with intractableepilepsy24 We assessed the pSer212LeufsTer9 variant aspathogenic according to the ACMGAMP guidelines byassigning the following criteria as evidence of pathogenicityPVS1 (null variant in a gene where loss of function is a knownmechanism of disease) PS2 (de novo variant in a patient withthe disease and no family history applied due to proven so-matic mosaicism) and PM2 (absence from large-scale controlpopulation databases)15

Deep targeted sequencing of SLC35A2confirms a wide range of VAFs in patientbrain tissueWe used 2 different targeted sequencing approaches to con-firm the SLC35A2 variant in DNA from 6 resected brainregions and to produce high-depth sequencing data at thevariant site as a means of correlating the prevalence of thevariant in the cells obtained from each region First we useda PCR-based approach (libricon) to selectively amplifya 100 bp region that included the variant of interest andconstructed a library from the resulting amplicons Our sec-ond variant confirmation approach included a custom directsequencing panel NND designed to capture and sequence allcoding regions of the SLC35A2 gene Both targetedapproaches confirmed the presence of the SLC35A2 variant atvarying proportions in different brain tissues (figure 3A)

The NND approach revealed variant read percentages rang-ing from 42 to 195 which were highly similar to ourinitial exome sequencing allele percentages (figure 3A) NNDuses UMIs added before PCR amplification permitting theidentification and removal of PCR duplicates after sequenc-ing Therefore on the basis of these assay attributes we choseNND as the methodology to confirm the SLC35A2 variant in6 additional brain tissues which had been archived in FFPEblocks from the patientrsquos initial surgery The variant waspresent in all 6 additional brain tissues at a range of percen-tages (figure 3B) Overall the hippocampus tissue (section C)had the highest proportion of the SLC35A2 variant allelewhereas portions of the occipital lobe (sections D2 and F)were among the lowest proportionsTa

ble

1Variantallele

frac

tionsan

dan

notationofso

maticSLC3

5A2c63

4_63

5delTC

varian

tiden

tifie

din

thepatientbrain

Chro

mPosition(hg1

9)Gene

HGVS

Matchedblood

Tempora

llobe

A1

Tempora

llobe

A2

Amyg

dala

BHippoca

mpusC

Occ

ipitallobe

D1

Occ

ipitallobe

D2

Depth

VAF

Depth

VAF

Depth

VAF

Depth

VAF

Depth

VAF

Depth

VAF

Depth

VAF

X48

7625

50SLC35

A2NM_0

0566

03c634

_635

delTC

pSer21

2Leu

fsTe

r910

8000

13

5300

18

4330

18

412

50

206

2770

153

980

16

6060

Abbreviations

Chrom

=ch

romoso

me

HGVS=Human

Gen

omeVariationSo

cietystan

dardnomen

clatureV

AF=va

rian

tallele

frac

tion

Gen

omicco

ord

inates

reflec

thuman

reference

genomebuild

GRCh37


Our initial transcriptome sequencing of total RNA yieldedlow read depth at the SLC35A2 c634_635delTC variant siteIn addition the expression level of SLC35A2 in all brain tis-sues was very low FPKM values for SLC35A2 expressionranged from 32 to 39 across all tissues As such we wereunable to correlate variant expression from transcriptomesequencing to the observed variability in allele proportionsfrom the DNA data Therefore we performed deep se-quencing using the NND targeted panel on brain-extractedRNA from 6 of the brain regions to determine whether thec634_635delTC variant is expressed Average read depth inRNA at the variant site was 5206x for all tissues and the VAFvaried from 15 to 77 demonstrating much lower pro-portions than were detected from DNA sequencing (figure3C) There was no clear correlation between DNA and RNAVAFs among all tissues

Somatic SLC35A2 VAF correlates with severityof electrophysiologic and radiographic findingsThe SLC35A2 VAF based on high-depth NND targetedsequencing of 12 brain regions obtained at surgery rangedfrom 42 (occipital lobe section D2) to 195 (hippo-campus section C) As shown previously in figure 2

intracranial EEG was consistent with seizure onset from thehippocampus which had the highest VAF but showed onlyoccasional seizure onset from the occipital lobe one of thelowest VAFs This observation prompted us to quantifyinterictal epileptiform discharges (spikes) for all involvedtissues (table 2) Average spikes per minute a standardmeasure of the severity of abnormal neurophysiologic ac-tivity correlated with the proportions of SLC35A2 variantallele in these different brain tissues (Pearson r = 0881 p =00087) with the highest rate of interictal epileptiform dis-charges (328 spikesmin) observed in the tissue with thehighest VAF (hippocampus 195) In addition as shown intable 2 the trend in correlation between VAFs and IEDactivity is reflected in a binary descriptor of seizure onsetrelative to brain region and to imaging-based analyses fromMRI and FDG-PET neuroimaging modalities Electrophys-iologic and neuroimaging findings generally correlated withhigher VAF compared with tissues with a lower VAF sug-gesting that the presence of the somatic mosaic SLC35A2variant is contributing to the seizure phenotype We com-pare these clinical findings with findings from a previouslypublished patient with the same SLC35A2 somatic variant intable 3

Figure 3 SLC35A2 c634_635delTC VAFs among different sequencing methods

(A) Comparison of SLC35A2 c634_635delTC VAFs in 6 brain tissues and blood comparator from the affected patient using 3 different sequencing techniquesAverage sequencing depth at the SLC35A2 variant position for each method is shown in parentheses in legend The apparent presence of the variant at lt1VAF in the blood sample for the libricon approach is likely due to the high error rate of this PCR-based approach and the high read depth achieved duringsequencing (B) VAFs in 12 brain and 1matched blood tissue using the NND SLC35A2 targeted sequencing kit (C) Comparison of VAFs in DNA and RNA from 6different brain tissues using the NND SLC35A2 targeted sequencing kit Only nucleic acids from high-quality frozen tissue samples were used for targetedRNA sequencing Average sequencing depth for each method is shown in parentheses in legends NND = NEBNext Direct VAF = variant allele fraction


DiscussionOur comparative analysis of electrophysiologic radiographichistopathologic and genomic data from multiple brain regionsfollowing left temporo-parieto-occipital resection in a patientwith West syndrome has illustrated a novel finding We iden-tified a somatic SLC35A2 variant pSer212LeufsTer9 whichwas detected initially by exome sequencing (VAF06ndash277) and confirmed with targeted gene sequencing(VAF 42ndash195) in various affected brain tissues but not in

blood This same variant was previously reported as a somaticvariant in the brain of an individual with FCD Ia24 Similar tothe previously published patient brain specimens harboring themosaic SLC35A2 variant from our patient showed corticaldyslamination FCD type 1c suggesting that the somatic vari-ant is associated with pathogenesis A germline de novo variant(pTyr145ProfsTer76) which results in truncation at the sameamino acid (p221) as our variant was reported in a patientwith EOEE and West syndrome25 The authors demonstrated

Table 2 Clinical phenotypic observations of all brain tissues

Hippocampus

Superiortemporalgyrus Amygdala

Inferiorparietalcortex

Temporallobe

Superioroccipital gyrus

Inferioroccipitalgyrus

Medialoccipitallobe

Section_ID C E B H1H2 A1A2 F D1D2 G1G2

Spikes perminute

328 253 243 188 138 137 57 na

Seizureonset

Yes Yes Yes Yes Yes No No na

MRI Abnormal Abnormal Abnormal Normal Abnormal Normal Normal Normal

FDG-PET Abnormal Abnormal Abnormal Abnormal Abnormal Normal Normal Normal

DNA VAF(NND)

195 111 113 125ndash143 48ndash64 48 42ndash65 148ndash164

RNA VAF(NND)

35 na 22 na 15ndash42 na 24ndash77 na

DNA VAF(exome)


Abbreviations FDG = fluorodeoxyglucose na = not applicable NND = NEBNext Direct VAF = variant allele fractionVarious clinical observations are listed for each brain tissue alongside the corresponding SLC35A2 c634_635delTC VAFs detected in each specimen

Table 3 Comparison of clinical features in our patient compared with a previously published patient with identical brainsomatic SLC35A2 c634_635delTC variant

Our patient Published case25

Variant pSer212LeufsTer9 (somatic) pSer212LeufsTer9 (somatic)

VAF Somatic 06ndash277 (exome sequencing) 42ndash195 (targeted sequencing) Somatic 143 (exome sequencing) 68(targeted sequencing)

Diagnosisseizure type

West syndrome Focal epilepsy

Age at onsetsex Male9 mo Male17 y

PreoperativeMRI Abnormal temporal lobe (underrepresentative of the dysplasia) Normal

Pathology FCD1c with varying degrees of cortical dyslamination and some gyri ldquosmall withcomplex configurationsrdquo

FCD1a hypercellular

HypsarrhythmiaEIEE

Yes No

Location Left posterior temporal-occipital lobe (nondominant) Right posterior temporal-occipital lobe(nondominant)

Abbreviations EIEE = early infantile epileptic encephalopathy FCD = focal cortical dysplasia VAF = variant allele fractionVarious clinical observations are listed for our patient and for a previously published patient who was reported to have the same somatic variant as ourpatient25


through in vitro cell culture studies using a mouse cell line thatthe pTyr145ProfsTer76 mutant expression was severely di-minished A comparison of SLC35A2 expression in brain tis-sues from our patient revealed that expression is indeeddiminished compared with brain tissues from other patientswith epilepsy (figure e-2 linkslwwcomNXGA273) Like-wise results from targeted sequencing in our patient affectedbrain tissues revealed the variant was estimated at much lowerread percentage in RNA than in DNA suggesting a nonsense-mediated decay (NMD) mechanism Clear correlation be-tween DNA variant allele and RNA VAFs or RNA expressionwithin each tissue was not observed However NMD is a dy-namic process and the extracted and sequenced RNA repre-sents only a snapshot of what is occurring in cells at any giventime point Therefore NMD is likely the reason for low ex-pression of the variant allele in RNA and lower expression ofSLC35A2 transcripts compared with other patients We predictthat overall SLC35A2 protein levels would be diminished in thepatient brain and therefore would associate with pathogenesisand the patient phenotype

The SLC35A2 gene encodes for a protein which transportsuridine-diphosphate-galactose from the cytosol to the Golgiwhere it functions in asparagine (N-linked) glycosylation anessential post-translational modification of proteins Studiesinmice revealed genetic inactivation of key components of theN-glycosylation pathway resulted in various neurologicdefects2627 Recessive conditions involving glycosylationdefects in humans (eg PMM2 ALG6 and SRD5A3) lead toa variety of clinical features most often neurologicsymptoms28ndash30 Individuals born with inherited geneticdefects in the N-glycosylation pathway often present withsevere neurologic abnormalities including seizures epilepticencephalopathy and cerebellar hypoplasia In particularheterozygous variants in the X-linked SLC35A2 in females areassociated with congenital disorder of glycosylation Type IImand are characterized by infantile-onset seizures hypsar-rhythmia severe intellectual disability and hypotonia253132

In addition to inherited germline variants previous studieshave reported similar to our patient identification of mosaicvariants in SLC35A2 as a cause of early-onset epileptic en-cephalopathy FCD and nonlesional focal epilepsy243233 Aplot of all pathogenic and likely pathogenic germline andsomatic SLC35A2 coding variants reported in ClinVar isshown in figure e-3 linkslwwcomNXGA273

Our group used an integrated combination of conventionalapproaches (radiology histopathology and electrophysiol-ogy) plus high-depth NGS to comprehensively analyze mul-tiple brain tissues from a single patient with intractableepilepsy Our sequencing results reveal that the SLC35A2variant affects RNA expression and is therefore likely associ-ated with the patient phenotype In concordance with thisconclusion the proportionality of the mosaicism of theSLC35A2 variant in various brain regions correlated with theseverity of neurophysiologic and presence or absence of ra-diographic abnormalities Although low prevalence of the

SLC35A2 variant allele still was sufficient to cause FCD pa-thology some tissues with low-level variant fractions appearedto be unaffected according to neuroimaging and onset of seizureactivity suggesting a potential threshold effect for epilepto-genesis Furthermore the amygdala and hippocampus did nothave apparent histologic pathology on initial examination Yetthe hippocampus and amygdala regions were determined to beamong the most epileptogenic regions in this patient high-lighting the strength of our study to verify underlying focalepileptogenic abnormalities using molecular pathology(ie DNA sequencing) in histologically normal tissue Collec-tively our findings support the understudied association of brainsomatic mosaicism with intractable epilepsy and reinforce theneed for studies that use a combination of clinical data analysisalongside deep NGS of affected brain tissue as a means toidentify underlying aspects of genetic pathogenicity

AcknowledgmentThe authors thank the patient and his family for participationin this study

Study fundingThis study was generously supported by funding from theNationwide Foundation Pediatric Innovation Fund

DisclosureKE Miller DC Koboldt KM Schieffer TA Bedrosian ECrist A Sheline and K Leraas report no disclosures VMagrini is a member of New England BioLabs Inc keyopinion leader group H Zhong P Brennan J Bush J FitchN Bir AR Miller CE Cottrell J Leonard JA Pindrik J ARusin S H Shah P White RK Wilson ER Mardis CRPierson and AP Ostendorf report no disclosures Go toNeurologyorgNG for full disclosures

Publication historyReceived by Neurology Genetics December 5 2019 Accepted in finalform May 5 2020

Appendix Authors

Name Location Contribution

Katherine EMiller PhD

Nationwide ChildrenrsquosHospital Columbus OH

Design andconceptualization of thestudy analysis andinterpretation of genomicdata and drafted the initialmanuscript

Daniel CKoboldt MS

Nationwide ChildrenrsquosHospital The Ohio StateUniversity Columbus OH

Design andconceptualization of thestudy analysis andinterpretation of genomicdata and revised themanuscript for intellectualcontent

Kathleen MSchiefferPhD


Analysis and interpretationof genomic data andrevised the manuscript forintellectual content

Continued


References1 Carvill GL Regan BM Yendle SC et al GRIN2A mutations cause epilepsy-aphasia

spectrum disorders Nat Genet 2013451073ndash10762 Claes L Del-Favero J Ceulemans B Lagae L Van Broeckhoven C De Jonghe P De

novo mutations in the sodium-channel gene SCN1A cause severe myoclonic epilepsyof infancy Am J Hum Genet 2001681327ndash1332

3 Singh NA Charlier C Stauffer D et al A novel potassium channel gene KCNQ2 ismutated in an inherited epilepsy of newborns Nat Genet 19981825ndash29

4 Schuele SU Luders HO Intractable epilepsy management and therapeutic alter-natives Lancet Neurol 20087514ndash524

5 Dwivedi R Ramanujam B Chandra PS et al Surgery for drug-resistant epilepsy inchildren N Engl J Med 20173771639ndash1647

6 Kwan P Brodie MJ Early identification of refractory epilepsy N Engl J Med 2000342314ndash319

7 Lee JH Huynh M Silhavy JL et al De novo somatic mutations in components of thePI3K-AKT3-mTOR pathway cause hemimegalencephaly Nat Genet 201244941ndash945

8 Moller RS Weckhuysen S Chipaux M et al Germline and somatic mutations in theMTOR gene in focal cortical dysplasia and epilepsy Neurol Genet 20162e118

9 Lim JS Kim WI Kang HC et al Brain somatic mutations in MTOR cause focalcortical dysplasia type II leading to intractable epilepsy Nat Med 201521395ndash400

10 Dobin A Davis CA Schlesinger F et al STAR ultrafast universal RNA-seq alignerBioinformatics 20132915ndash21

11 Blumcke I ThomM Aronica E et al The clinicopathologic spectrum of focal corticaldysplasias a consensus classification proposed by an ad hoc Task Force of the ILAEDiagnostic Methods Commission Epilepsia 201152158ndash174

12 Kelly BJ Fitch JR Hu Y et al Churchill an ultra-fast deterministic highly scalableand balanced parallelization strategy for the discovery of human genetic variation inclinical and population-scale genomics Genome Biol 2015166

13 DePristo MA Banks E Poplin R et al A framework for variation discovery andgenotyping using next-generation DNA sequencing data Nat Genet 201143491ndash498

14 Cibulskis K Lawrence MS Carter SL et al Sensitive detection of somatic pointmutations in impure and heterogeneous cancer samples Nat Biotechnol 201331213ndash219

15 Richards S Aziz N Bale S et al Standards and guidelines for the interpretation ofsequence variants a joint consensus recommendation of the American College ofMedical Genetics and Genomics and the Association for Molecular Pathology GenetMed 201517405ndash424

16 LoveMI Huber W Anders S Moderated estimation of fold change and dispersion forRNA-seq data with DESeq2 Genome Biol 201415550

17 Emerman AB Bowman SK Barry A et al NEBNext direct a novel rapidhybridization-based approach for the capture and library conversion of genomicregions of interest Curr Protoc Mol Biol 20171197 30 31ndash37 30 24

18 Koboldt DC Zhang Q Larson DE et al VarScan 2 somatic mutation and copynumber alteration discovery in cancer by exome sequencing Genome Res 201222568ndash576

Appendix (continued)


Tracy ABedrosianPhD


Analysis and interpretationof genomic data andrevised the manuscript

Erin CristMMSc LGC


Major role in acquisition ofdata

AdrienneShelineMPH



KristenLeraas MS



VincentMagriniPhD



HuachunZhong MS



PatrickBrennanMS



JocelynBush MS



James FitchBS



Natalie BirBS



Anthony RMiller PhD



Catherine ECottrellPhD



JeffreyLeonardMD



Jonathan APindrik MD



Jerome ARusin MD


Major role in acquisition ofdata and analysis andinterpretation of clinicaldata

Summit HShah MDMPH



Peter WhitePhD


Design andconceptualization of thestudy major role inacquisition of data andanalysis and interpretationof genomic data



Richard KWilson PhD


Design andconceptualization of thestudy

Elaine RMardis PhD



ChristopherR PiersonMD PhD


Major role in acquisition ofdata analysis andinterpretation of clinicaldata and revised themanuscript for intellectualcontent

Adam POstendorfMD


Major role in acquisition ofdata analysis andinterpretation of clinicaldata and drafted the initialmanuscript


19 Li H A statistical framework for SNP calling mutation discovery association mappingand population genetical parameter estimation from sequencing data Bioinformatics2011272987ndash2993

20 Wirrell EC Shellhaas RA Joshi C et al How should children with West syndrome beefficiently and accurately investigated Results from the National Infantile SpasmsConsortium Epilepsia 201556617ndash625

21 Poirier K Lebrun N Broix L et al Mutations in TUBG1 DYNC1H1 KIF5C andKIF2A cause malformations of cortical development and microcephaly Nat Genet201345639ndash647

22 Lin Z Liu Z Li X et al Whole-exome sequencing identifies a novel de novo mutationin DYNC1H1 in epileptic encephalopathies Sci Rep 20177258

23 Lek M Karczewski KJ Minikel EV et al Analysis of protein-coding genetic variationin 60706 humans Nature 2016536285ndash291

24 Winawer MR GriffinNG Samanamud J et al Somatic SLC35A2 variants in the brainare associated with intractable neocortical epilepsy Ann Neurol 2018831133ndash1146

25 Kodera H Nakamura K Osaka H et al De novo mutations in SLC35A2 encodinga UDP-galactose transporter cause early-onset epileptic encephalopathy Hum Mutat2013341708ndash1714

26 Ye Z Marth JD N-glycan branching requirement in neuronal and postnatal viabilityGlycobiology 200414547ndash558

27 Thiel C Korner C Mouse models for congenital disorders of glycosylation J InheritMetab Dis 201134879ndash889

28 Cantagrel V Lefeber DJ Ng BG et al SRD5A3 is required for converting polyprenol todolichol and is mutated in a congenital glycosylation disorder Cell 2010142203ndash217

29 Imbach T Grunewald S Schenk B et al Multi-allelic origin of congenital disorder ofglycosylation (CDG)-Ic Hum Genet 2000106538ndash545

30 Freeze HH Eklund EA Ng BG Patterson MC Neurology of inherited glycosylationdisorders Lancet Neurol 201211453ndash466

31 Kimizu T Takahashi Y Oboshi T et al A case of early onset epileptic encephalopathywith de novo mutation in SLC35A2 clinical features and treatment for epilepsy BrainDev 201739256ndash260

32 Ng BG Buckingham KJ Raymond K et al Mosaicism of the UDP-galactose trans-porter SLC35A2 causes a congenital disorder of glycosylation Am J Hum Genet201392632ndash636

33 Sim NS Seo Y Lim JS et al Brain somatic mutations in SLC35A2 cause intractableepilepsy with aberrant N-glycosylation Neurol Genet 20184e294


DOI 101212NXG000000000000046020206 Neurol Genet

Katherine E Miller Daniel C Koboldt Kathleen M Schieffer et al mosaicism correlates with clinical findings in epilepsy brain tissueSLC35A2Somatic

This information is current as of June 17 2020

reserved Online ISSN 2376-7839Published by Wolters Kluwer Health Inc on behalf of the American Academy of Neurology All rightsan open-access online-only continuous publication journal Copyright Copyright copy 2020 The Author(s)

is an official journal of the American Academy of Neurology Published since April 2015 it isNeurol Genet

ServicesUpdated Information amp

httpngneurologyorgcontent64e460fullhtmlincluding high resolution figures can be found at

References httpngneurologyorgcontent64e460fullhtmlref-list-1

This article cites 33 articles 3 of which you can access for free at

Citations httpngneurologyorgcontent64e460fullhtmlotherarticles

This article has been cited by 4 HighWire-hosted articles

Subspecialty Collections

httpngneurologyorgcgicollectioninfantile_spasmsInfantile spasms

httpngneurologyorgcgicollectionepilepsy_surgery_Epilepsy surgery

httpngneurologyorgcgicollectioncortical_dysplasiaCortical dysplasia

httpngneurologyorgcgicollectionall_geneticsAll Geneticsfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpngneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpngneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online



Many genetic epilepsies are characterized by Mendelian in-heritance notably in ion channel genes (eg SCN12AKCNQ2 and GRIN12A2B2D) which also may occur denovo in individuals without any family history of epilepticdisorders1ndash3 Approximately one-third of epilepsies aretreatment refractory due to failure of medication to alleviateor ameliorate seizures4ndash6 In select cases electrophysiologicbrain mapping can identify affected regions and the re-section or disconnection of these regions is potentially cu-rative For some patients the likelihood of becomingseizure-free after surgical resection or disconnection is ashigh as 8056

Several studies have recently identified somatic mosaicismassociated with epilepsy in which genetic variants are presentonly in a subset of brain cells identified in resected tissue Themost robust association involves genes in the PI3K-AKT-mTOR pathway wherein hyperactivation of the mTORpathway occurs in patients with hemimegalencephaly andfocal cortical dysplasia (FCD)7ndash10 Still the genetic etiology ofmany epileptic phenotypes associated with somatic mosai-cism remains ill defined

At Nationwide Childrenrsquos Hospital a National Associationof Epilepsy Centers Level 4 epilepsy center we have estab-lished a research program to identify brain somatic mosai-cism in patients with treatment refractory epilepsy who haveundergone surgical resection (figure e-1 linkslwwcomNXGA273) High-depth next-generation sequencing(NGS) is performed on brain DNA in comparison to blood-derived DNA Somatic variants are assessed for pathoge-nicity and analyzed in the context of clinically obtainedmultimodal information We hypothesize that tissue-specific somatic mosaicism represents an important ge-netic etiology in epilepsy

MethodsStandard protocol approvals registrationsand patient consentsThis study was designed to perform genomic profiling ofexcised tissue and matched (blood) samples from a patientwith refractory epilepsy Written informed consent wasobtained and the patient was enrolled as part of aninstitutional review board-approved research study (IRB18-00786) within The Steve and Cindy Rasmussen Institute forGenomic Medicine at Nationwide Childrenrsquos Hospital

Data availabilityThe SLC35A2 variant information was deposited inClinVar (ncbinlmnihgovclinvar) under accessionnumber VCV0006351171

HistopathologyHematoxylin and eosin staining was performed as part ofroutine clinical workup according to recommendations of theInternational League Against Epilepsy11

NeurophysiologyThe index patient underwent multiple extracranial EEG anda single intracranial EEG All data were digitally acquired andinterpreted with a Nihon Kohden system Interpretationswere performed by a board-certified clinical neurophysiologistaccording to standard clinical care Quantification of epilep-tiform activity was performed by counting spikes polyspikesand ripples during a 5-minute sample at the beginning of eachhour during the last day of recording (120 total minutes from24 epochs) For each tissue the correlation between activity(average spikes per minute) and variant allele fraction (VAF)(average tissue VAF from NEBNext Direct [NND]) wascalculated in R using the cortest operation

MRIThe index patient was imaged 3 times before the final diagnosisThe first 2 scans were obtained on a GE Discovery MR750 3Tmagnet (GE Medical Systems Milwaukee WI) using a 32-channel head coil with sedation performed by a pediatric anes-thesiologist The first 2 scans included 3 plane brain volumeimaging long inversion recovery (T1 volumetric) axialcoronalT2 FSE axial susceptibility weighted and axial diffusionsequences The second scan also included 3 plane T2 fluid-attenuated inversion recovery (FLAIR) CUBE (T2 volumetric)axial diffusion tensor imaging using 30 gradient encodingdirections and postgadobutrol (Gadovist Bayer HealthCareWhippany NJ) coronal T2 FLAIR 3 plane brain volume im-aging long inversion recovery (T1 volumetric) and axial T1 spinecho sequences The third MRI scan was performed on a Sie-mens PRISMA 3Tmagnet (Siemens Erlangen Germany) usinga 32-channel head coil with sedation by a pediatric anesthesi-ologist The third scan included the following 3 plane magne-tization prepared-rapid gradient ECHO (T1 volumetric) 3plane T2 FLAIR sampling perfection with application-optimized(T2 volumetric) axial diffusion tensor imaging using 30 gradientencoding directions axial and coronal T2 high resolution axialsusceptibility-weighted imaging and postgadobutrol (GadovistBayer HealthCare Whippany NJ) coronal T2 FLAIR 3 plane

GlossaryACMG = American College of Medical Genetics and Genomics AMP = Association for Molecular Pathology FCD = focalcortical dysplasia FDG = fluorodeoxyglucose FFPE = formalin-fixed paraffin embedded FPKM = Fragments Per Kilobase oftranscript per Million IGV = Integrated Genomics Viewer NGS = next-generation sequencing NMD = nonsense-mediateddecay NND = NEBNext Direct UMI = unique molecular identifier VAF = variant allele fraction VUS = variant of unknownsignificance



























ble

1Variantallele

frac

tionsan

dan

notationofso

maticSLC3

5A2c63

4_63

5delTC

varian

tiden

tifie

din

thepatientbrain

Chro

mPosition(hg1

9)Gene

HGVS

Matchedblood

Tempora

llobe

A1

Tempora

llobe

A2

Amyg

dala

BHippoca

mpusC

Occ

ipitallobe

D1

Occ

ipitallobe

D2

Depth

VAF

Depth

VAF

Depth

VAF

Depth

VAF

Depth

VAF

Depth

VAF

Depth

VAF

X48

7625

50SLC35

A2NM_0

0566

03c634

_635

delTC

pSer21

2Leu

fsTe

r910

8000

13

5300

18

4330

18

412

50

206

2770

153

980

16

6060

Abbreviations

Chrom

=ch

romoso

me

HGVS=Human

Gen

omeVariationSo

cietystan

dardnomen

clatureV

AF=va

rian

tallele

frac

tion

Gen

omicco

ord

inates

reflec

thuman

reference

genomebuild

GRCh37











Hippocampus



Temporallobe



Medialoccipitallobe


Spikes perminute

328 253 243 188 138 137 57 na

Seizureonset




DNA VAF(NND)


RNA VAF(NND)


DNA VAF(exome)












FCD1a hypercellular

HypsarrhythmiaEIEE

Yes No













Appendix Authors





Daniel CKoboldt MS






Continued























Tracy ABedrosianPhD



Erin CristMMSc LGC



AdrienneShelineMPH



KristenLeraas MS



VincentMagriniPhD



HuachunZhong MS



PatrickBrennanMS



JocelynBush MS



James FitchBS



Natalie BirBS



Anthony RMiller PhD






JeffreyLeonardMD






Jerome ARusin MD



Summit HShah MDMPH



Peter WhitePhD





Richard KWilson PhD



Elaine RMardis PhD






Adam POstendorfMD






































Reprints





























ble

1Variantallele

frac

tionsan

dan

notationofso

maticSLC3

5A2c63

4_63

5delTC

varian

tiden

tifie

din

thepatientbrain

Chro

mPosition(hg1

9)Gene

HGVS

Matchedblood

Tempora

llobe

A1

Tempora

llobe

A2

Amyg

dala

BHippoca

mpusC

Occ

ipitallobe

D1

Occ

ipitallobe

D2

Depth

VAF

Depth

VAF

Depth

VAF

Depth

VAF

Depth

VAF

Depth

VAF

Depth

VAF

X48

7625

50SLC35

A2NM_0

0566

03c634

_635

delTC

pSer21

2Leu

fsTe

r910

8000

13

5300

18

4330

18

412

50

206

2770

153

980

16

6060

Abbreviations

Chrom

=ch

romoso

me

HGVS=Human

Gen

omeVariationSo

cietystan

dardnomen

clatureV

AF=va

rian

tallele

frac

tion

Gen

omicco

ord

inates

reflec

thuman

reference

genomebuild

GRCh37











Hippocampus



Temporallobe



Medialoccipitallobe


Spikes perminute

328 253 243 188 138 137 57 na

Seizureonset




DNA VAF(NND)


RNA VAF(NND)


DNA VAF(exome)












FCD1a hypercellular

HypsarrhythmiaEIEE

Yes No













Appendix Authors





Daniel CKoboldt MS






Continued























Tracy ABedrosianPhD



Erin CristMMSc LGC



AdrienneShelineMPH



KristenLeraas MS



VincentMagriniPhD



HuachunZhong MS



PatrickBrennanMS



JocelynBush MS



James FitchBS



Natalie BirBS



Anthony RMiller PhD






JeffreyLeonardMD






Jerome ARusin MD



Summit HShah MDMPH



Peter WhitePhD





Richard KWilson PhD



Elaine RMardis PhD






Adam POstendorfMD






































Reprints













Hippocampus



Temporallobe



Medialoccipitallobe


Spikes perminute

328 253 243 188 138 137 57 na

Seizureonset




DNA VAF(NND)


RNA VAF(NND)


DNA VAF(exome)












FCD1a hypercellular

HypsarrhythmiaEIEE

Yes No













Appendix Authors





Daniel CKoboldt MS






Continued























Tracy ABedrosianPhD



Erin CristMMSc LGC



AdrienneShelineMPH



KristenLeraas MS



VincentMagriniPhD



HuachunZhong MS



PatrickBrennanMS



JocelynBush MS



James FitchBS



Natalie BirBS



Anthony RMiller PhD






JeffreyLeonardMD






Jerome ARusin MD



Summit HShah MDMPH



Peter WhitePhD





Richard KWilson PhD



Elaine RMardis PhD






Adam POstendorfMD






































Reprints













Appendix Authors





Daniel CKoboldt MS






Continued























Tracy ABedrosianPhD



Erin CristMMSc LGC



AdrienneShelineMPH



KristenLeraas MS



VincentMagriniPhD



HuachunZhong MS



PatrickBrennanMS



JocelynBush MS



James FitchBS



Natalie BirBS



Anthony RMiller PhD






JeffreyLeonardMD






Jerome ARusin MD



Summit HShah MDMPH



Peter WhitePhD





Richard KWilson PhD



Elaine RMardis PhD






Adam POstendorfMD






































Reprints

























Tracy ABedrosianPhD



Erin CristMMSc LGC



AdrienneShelineMPH



KristenLeraas MS



VincentMagriniPhD



HuachunZhong MS



PatrickBrennanMS



JocelynBush MS



James FitchBS



Natalie BirBS



Anthony RMiller PhD






JeffreyLeonardMD






Jerome ARusin MD



Summit HShah MDMPH



Peter WhitePhD





Richard KWilson PhD



Elaine RMardis PhD






Adam POstendorfMD






































Reprints






































Reprints




somatic slc35a2 mosaicism correlates with clinical ﬁ

Documents