Informacije

Šta predstavljaju ocjene haploinsuficijencije u ClinVar bazi podataka?

Šta predstavljaju ocjene haploinsuficijencije u ClinVar bazi podataka?


We are searching data for your request:

Forums and discussions:
Manuals and reference books:
Data from registers:
Wait the end of the search in all databases.
Upon completion, a link will appear to access the found materials.

Razumijem da se haploinsuficijencija javlja kada je jedna od dvije kopije gena mutirana do te mjere da je neupotrebljiva, a preostala pojedinačna kopija nije u stanju nositi se sa zahtjevima stanice za datim peptidom, što rezultira fenotipskim nedostatkom.

U bazi podataka clinVar postoji "skor haploinsuficijencije" (ovde primjer) koji se čini da se kreće od 0 do 3 (iako je to samo anegdota; na osnovu gena koje sam pogledao). Pretpostavljam da se odnosi na to koliko je ćelija fenotipska ili koliko je manjkava. Međutim, ne uspijevam pronaći mjesto gdje je to jasno naznačeno ...

Da li sam u pravu u ovoj pretpostavci, i ako je tako, šta predstavlja ocjena 3 (na primjer)?


U redu, konačno sam pronašao službeni izvor koji opisuje šta predstavljaju ocjene osjetljivosti na dozu.

0 ne predstavlja dokaz osjetljivosti na dozu: kliničko tumačenje je da su razlike u dozama "vjerojatno benigne".

1 predstavlja malo dokaza o osjetljivosti na dozu: Ovo se može označiti ako samo sekundarni dokazi pokazuju osjetljivost na dozu; klinička interpretacija je da je osjetljivost na dozu "neizvjesna".

2 predstavlja "dokaze u nastajanju", što znači da dokazi prilično snažno sugeriraju da je gen osjetljiv na dozu, ali da nema dovoljno dokaza koji bi to dokazali. Kliničko tumačenje je da je osjetljivost na dozu "nesigurna, vjerovatno patogena".

3 predstavlja "Dovoljan dokaz", što znači da su dokazi dovoljni da se dokaže da je gen osjetljiv na dozu. Klinička interpretacija je "patogena".


Izvor:

  • Riggs ER i sur., Ka procesu zasnovanom na dokazima za kliničku interpretaciju varijacije broja kopija. Clin Genet. 81 (5): 403-412.

Ponovno razmatranje morbidnog genoma Mendelovih poremećaja

Patogenost mnogih Mendelovih varijanti dovedena je u pitanje velikim naporima za sekvencioniranje. Međutim, mnoge rijetke i benigne "mutacije bolesti" teško je analizirati zbog njihove rijetkosti. Variome Saudijske Arabije obogaćen je za homozigotnost zbog inbreedinga, što je ključna prednost koja se može iskoristiti za kritičko ispitivanje ranije objavljenih varijanti.

Rezultati

Uporedili smo sve „mutacije povezane sa bolešću“ navedene u bazi podataka o mutacijama ljudskih gena (HGMD) i ClinVar-u, uključujući „varijante neizvjesnog značaja“ (VOUS). Otkrivamo da upotreba javnih baza podataka, uključujući 1000 genoma, ExAC i Kaviar, može mnoge od ovih varijanti preklasificirati u vjerovatno benigne. Naš Saudijski program ljudskog genoma (SHGP) može reklasificirati mnoge varijante koje su rijetke u javnim bazama podataka. Nadalje, SGPD nam omogućava da promatramo mnoge ranije prijavljene varijante u homozigotnom stanju, a naša opsežna fenotipizacija učesnika omogućava da se pokaže nedostatak fenotipa za ove varijante, čime se dovodi u pitanje njihova patogenost uprkos njihovoj rijetkosti. Takođe nalazimo da 18 VOUS BRCA1 i BRCA2 varijante navedene u BRCA Exchange prisutne su barem jednom u homozigotnom stanju kod pacijenata koji nemaju obilježja Fanconijeve anemije. Ohrabrujuće je da bismo mogli recipročno pokazati da nijedan od onih koji su označeni kao „patogeni“ nije primijećen u homozigotnoj statui kod pojedinaca kojima nedostaje Fanconijev fenotip u našoj bazi podataka.

Zaključak

Naše istraživanje pokazuje važnost ponovnog posjeta bazama podataka o bolestima koristeći javne resurse, kao i resurse specifične za populaciju, kako bi se poboljšala specifičnost morbidnog genoma mendelskih bolesti kod ljudi.


ClinGen je 2013. godine osnovao Nacionalni institut za istraživanje humanog genoma, ClinGen je rastući zajednički napor koji uključuje tri bespovratna sredstva, devet glavnih istraživača i preko 1500 saradnika iz više od 35 zemalja. Ispod je niz nedavnih ažuriranja na kojima ClinGen radi.

Gledajte snimke i pridružite se forumu

Radna grupa ClinGen za prednike i raznolikost održava mjesečni Forum koji se odnosi na raznolikost u istraživanjima humane genetike i kliničkoj genomici.

Panel za karijeru kliničke genomike

Počevši od 30. juna, pridružite nam se na virtuelnoj seriji panela na kojoj učenici mogu čuti o različitim karijerama u genomici.

Edukacija o biokuratoru

Stranica Obrazovni materijali Biokuratora sadrži linkove na video zapise poziva Radne grupe Biokurator o različitim temama.

Volontirajte za kuriranje

Molimo vas da napravite kratku anketu kako biste nam rekli više o vašim interesima i željenom nivou uključenosti kako bismo vas mogli upariti s odgovarajućom aktivnosti kuriranja i/ili stručnim vijećem.


Bojkot, K. M. i dr. Međunarodna saradnja za dijagnosticiranje svih rijetkih genetskih bolesti. Am. J. Hum. Genet. 100, 695–705, https://doi.org/10.1016/j.ajhg.2017.04.003 (2017).

Retterer, K. et al. Klinička primjena sekvenciranja cijelog egzoma u svim kliničkim indikacijama. Genet. Med. 18, 696–704, https://doi.org/10.1038/gim.2015.148 (2016).

Trujillano, D. et al. Kliničko sekvenciranje egzoma: rezultati iz 2819 uzoraka koji odražavaju 1000 porodica. EUR. J. Hum. Genet. 25, 176–182, https://doi.org/10.1038/ejhg.2016.146 (2017).

Dillon, O. J. i sur. Sekvenciranje egzoma ima veći dijagnostički prinos u poređenju sa simuliranim panelima specifičnim za bolest kod djece sa sumnjom na monogenske poremećaje. EUR. J. Hum. Genet. 26, 644–651, https://doi.org/10.1038/s41431-018-0099-1 (2018).

Sobreira, N. L. M. i sur. Matchmaker Exchange. Curr. Protoc. Hum. Genet. 95, 9.31.1–9.31.15, https://doi.org/10.1002/cphg.50 (2017).

Au, P. Y. B. et al. GeneMatcher pomaže u identifikaciji novog sindroma malformacije s intelektualnim teškoćama, jedinstvenih facijalnih dismorfizama i abnormalnosti skeleta i vezivnog tkiva uzrokovanih de novo varijantama u HNRNPK. Hum. Mutat. 36, 1009–1014, https://doi.org/10.1002/humu.22837 (2015).

O’Donnell-Luria, A. H. i sur. Heterozigotne varijante KMT2E uzrokuju niz neurorazvojnih poremećaja i epilepsiju. Am. J. Hum. Genet. 104, 1210–1222, https://doi.org/10.1016/j.ajhg.2019.03.021 (2019).

Bruel, A.-L. et al. 2,5 godine iskustva u dijeljenju podataka GeneMatcher-a: moćno oruđe za identifikaciju novih gena odgovornih za rijetke bolesti. Genet. Med. 21, 1657–1661, https://doi.org/10.1038/s41436-018-0383-z (2019).

Leduc, M. S. et al. De novo prividne mutacije gubitka funkcije u PRR12 kod tri pacijenta s intelektualnim invaliditetom i abnormalnostima šarenice. Hum. Genet. 137, 257–264, https://doi.org/10.1007/s00439-018-1877-0 (2018).

Córdova-Fletes, C. i sur. De novo t(1019)(q22.3q13.33) dovodi do transkripta recipročne fuzije ZMIZ1/PRR12 kod djevojčice sa intelektualnim invaliditetom i neuropsihijatrijskim promjenama. Neurogenetika. 16, 287–298, https://doi.org/10.1007/s10048-015-0452-2 (2015).

Miller, J. A. et al. Transkripcijski pejzaž prenatalnog ljudskog mozga. Priroda. 508, 199–206, https://doi.org/10.1038/nature13185 (2014).

Bult, C. J., Blake, J. A., Smith, C. L., Kadin, J. A., Richardson, J. E. & amp. Mouse Genome Database Group. Baza genoma miša (MGD) 2019. Nukleinske kiseline Res. 47, D801 – D806, https://doi.org/10.1093/nar/gky1056 (2019).

Karczewski, K. J. et al. Spektar ograničenja mutacija kvantificiran je varijacijom u 141.456 ljudi. Priroda. 581, 434–443, https://doi.org/10.1038/s41586-020-2308-7 (2020).

Sobreira, N., Schiettecatte, F., Valle, D. & amp Hamosh, A. GeneMatcher: odgovarajući alat za povezivanje istraživača koji se zanimaju za isti gen. Hum. Mutat. 36, 1928–1930, https://doi.org/10.1002/humu.22844 (2015).

Gu, S. i sur. Mehanizmi za složene kromosomske insercije. PLoS Genet. 12, e1006446, https://doi.org/10.1371/journal.pgen.1006446 (2016).

Tapial, J. i sur. Atlas alternativnih profila spajanja i funkcionalnih asocijacija otkriva nove regulatorne programe i gene koji istovremeno izražavaju više glavnih izoforma. Genome Res. 27, 1759–1768, https://doi.org/10.1101/gr.220962.117 (2017).

Firth, H.V. et al. DECIFHER: Baza podataka hromozomske neravnoteže i fenotipa kod ljudi koristeći resurse ansambla. Am. J. Hum. Genet. 84, 524–533, https://doi.org/10.1016/j.ajhg.2009.03.010 (2009).

Mirzaa, G. i sur. PIK3CA-povezani razvojni poremećaji pokazuju različite klase mutacija s varijabilnom ekspresijom i raspodjelom tkiva. JCI Insight. 1, e87623 (2016).

Stolerman, E. S. i sar. Genetske varijante u genu KDM6B povezane su sa kašnjenjima u neurorazvoju i dismorfnim karakteristikama. Am. J. Med. Genet. A. 179, 1276–1286, https://doi.org/10.1002/ajmg.a.61173 (2019).

Unlu, G. i sur. GRIK5 genetski regulirana ekspresija povezana s očnim i vaskularnim fenomenima: otkriće putem iteracije među biobankama, elektroničkim zdravstvenim zapisima i zebricama. Am. J. Hum. Genet. 104, 503–519, https://doi.org/10.1016/j.ajhg.2019.01.017 (2019).

Reis, L. M. i sar. Dominantne varijante u PRR12 rezultiraju unilateralnom ili bilateralnom kompleksnom mikroftalmijom. Clin. Genet. https://doi.org/10.1111/cge.13897 (2020).

Fishilevich, S. i sur. GeneHancer: integracija pojačivača i ciljnih gena na čitavom genomu u GeneCards. Baza podataka (Oxford). 2017, bax028 (2017).

Zerbino, D. R. et al. Ansambl 2018. Nukleinske kiseline Res. 46, D754 – D761, https://doi.org/10.1093/nar/gkx1098 (2018).

Bernstein, B. E. et al. Dvovalentna struktura kromatina označava ključne razvojne gene u embrionalnim matičnim stanicama. Cell. 125, 315–326, https://doi.org/10.1016/j.cell.2006.02.041 (2006).

Harikumar, A. & amp Meshorer, E. Remodeliranje hromatina i bivalentne histonske modifikacije u embrionalnim matičnim ćelijama. EMBO Rep. 16, 1609–1619, https://doi.org/10.15252/embr.201541011 (2015).

Nagase, T. i sur. Predviđanje kodirajućih sekvenci neidentificiranih ljudskih gena. XV. Kompletne sekvence 100 novih cDNA klonova iz mozga koji kodiraju velike proteine ​​in vitro. DNA Res. 6, 337–345, https://doi.org/10.1093/dnares/6.5.337 (1999).

Havugimana, P. C. i sar. Popis kompleksa topljivih proteina u ljudi. Cell. 150, 1068–1081, https://doi.org/10.1016/j.cell.2012.08.011 (2012).

Kim, B. R. i sar. Identifikacija SOX2 interaktoma pomoću BioID-a otkriva EP300 kao posrednika u vezi sa SOX2-zavisnom skvamoznom diferencijacijom i rastu pločastih karcinoma pluća. Mol. Cell. Proteomika. 16, 1864–1888, https://doi.org/10.1074/mcp.M116.064451 (2017).

Giurato, G. et al. Kvantitativno mapiranje RNA-posredovanog nuklearnog estrogenskog receptora β interaktoma u ljudskim stanicama raka dojke. Sci Data. 5, 180031, https://doi.org/10.1038/sdata.2018.31 (2018).

Fountain, M. D. et al. Patogene varijante u USP7 uzrokuju neurorazvojni poremećaj sa kašnjenjem govora, promijenjenim ponašanjem i neurološkim anomalijama. Genet. Med. 21, 1797–1807 (2019).

Ragge, N.K. et al. SOX2 anoftalmijski sindrom. Am. J. Med. Genet. A. 135, 1–7, https://doi.org/10.1002/ajmg.a.30642 (2005). diskusija 8.

Baetens, D. et al. Bijalelne i monoalelne varijante ESR2 povezane s 46, XY poremećajima spolnog razvoja. Genet. Med. 20, 717–727, https://doi.org/10.1038/gim.2017.163 (2018).

Balci, T. B. i dr. Razotkrivanje Occamove britve: dijagnosticiranje višestrukih genetskih bolesti u porodicama sekvenciranjem cijelog egzoma. Clin. Genet. 92, 281–289, https://doi.org/10.1111/cge.12987 (2017).

Posey, J. E. i sur. Rezolucija fenotipova bolesti kao rezultat multilokusne genomske varijacije. N. Engl. J. Med. 376, 21–31, https://doi.org/10.1056/NEJMoa1516767 (2017).

de Geus, C. M. i sar. Smjernice za CHARGE sindrom i karika koja nedostaje: snimanje lobanje. Am. J. Med. Genet. C Semin. Med. Genet. 175, 450–464, https://doi.org/10.1002/ajmg.c.31593 (2017).

Andreou, A.M. et al. TBX22 missense mutacije pronađene kod pacijenata sa X-vezanim rascjepom nepca utječu na vezivanje DNK, sumoilaciju i transkripcijsku represiju. Am. J. Hum. Genet. 81, 700–712, https://doi.org/10.1086/521033 (2007).

Al-Baradie, R. et al. Sindrom radijalnih zraka Duane (Okihiro sindrom) preslikava se na 20q13 i rezultat je mutacija u SALL4, novom članu porodice SAL. Am. J. Hum. Genet. 71, 1195–1199, https://doi.org/10.1086/343821 (2002).

Obayashi, T., Kagaya, Y., Aoki, Y., Tadaka, S. & Kinoshita, K. COXPRESdb v7: baza podataka o koekspresiji gena za 11 životinjskih vrsta podržana od 23 platforme za koekspresiju za tehničku evaluaciju i evolucijsko zaključivanje. Nukleinske kiseline Res. 47, D55 – D62, https://doi.org/10.1093/nar/gky1155 (2019).

Jiao, X. i sur. DAVID-WS: web servis sa stanjem za olakšavanje analize liste gena/proteina. Bioinformatika. 28, 1805–1806, https://doi.org/10.1093/bioinformatics/bts251 (2012).

Choudhary, C. et al. Acetilacija lizina cilja na proteinske komplekse i ko-regulira glavne stanične funkcije. Nauka. 325, 834–840, https://doi.org/10.1126/science.1175371 (2009).


2 IZVJEŠTAJ O SLUČAJU

2.1 Urednička politika i etička razmatranja

2.1.1 Etičko odobrenje i pristanak za učešće

2.1.2 Pristanak pacijenata za objavljivanje

Informacije o pacijentu u ovom izvještaju o slučaju odobrio je staratelj pacijenta i on razumije sadržaj članka.

2.2 Klinički izvještaj

12-godišnji dječak je primljen u lokalnu bolnicu zbog povišene temperature i slabosti. Međutim, njegovo se stanje brzo pogoršalo te je prebačen na dječju jedinicu intenzivne njege kad mu se javilo teško disanje. Prestao je da diše 2 h kasnije, a uspostavljen je vještački disajni put sa mehaničkom ventilacijom. Međutim, još 2 sata kasnije, srce mu je iznenada stalo. Kardiopulmonalna reanimacija (CPR) nije bila uspješna. Pozvan je tim za ekstrakorporalnu membransku oksigenaciju (ECMO) u našoj bolnici, koji je prebačen u našu bolnicu pod ECMO podrškom na daljnje liječenje.

Dijete je bilo zdravo do svoje 2 godine. Njegova istorija rođenja bila je normalna. Njegovi roditelji nisu bili u srodstvu i bili su zdravi. Kad je imao 2,5 godine, hodao je zapanjujućim hodom i često je padao. Kao rezultat toga, dijete je odmah pregledano i zbrinuto u lokalnoj bolnici. Doktori su prvo izvršili rutinske biohemijske i fizičke preglede, a pregled je otkrio da je kod djeteta nestao refleks koljena. Elektromiografija (EKG) ukazuje na perifernu neuropatiju, a uglavnom se radi o oštećenju osjetnih živaca. Istovremeno, dijete je imalo simptome ponovljene mišićne slabosti. Stoga je djetetu dijagnosticiran Guillain -Barréov sindrom (GBS) (Hughes & Cornblath, 2005). Visoke doze imunoglobulina primjenjivane su 3 dana i stanje mu se poboljšalo. Mekobalamin, citicolin i adenozin trifosfat davani su dugo vremena kako bi mu se obnovili kranijalni živci. Međutim, ponovo je hospitaliziran zbog slabosti mišića kada je imao 6 godina i ponovo mu je dijagnosticiran GBS.

Tri mjeseca prije dolaska u našu bolnicu, dječak je umalo umro od iznenadnog srčanog zastoja. ECMO tim je spasio i stabilizovao njegovu cirkulaciju i disanje. Na kompjuterizovanoj tomografiji mozga i MRI predložene su lezije moždanog debla (slika 1). EEG pregled je pokazao da se pozadinska aktivnost usporila. Dijete je imalo lumbalnu punkciju. Pregled cerebrospinalne tečnosti (CSF) pokazao je normalne nivoe proteina (290,17 mg/L N: 150-450 mg/L) sa leukocitozom (WBC 14 × 10 6/L). I dijete je imalo simptome ataksije i oslabljene tetivne reflekse. Iako nije bilo fenomena razdvajanja proteina i stanica, pedijatar i neurolog su visoko sumnjali da ima najteži GBS-MFS (Miller-Fisherov sindrom). Dječak i njegovi roditelji prošli su kliničko sekvenciranje egzoma (CES) kako bi isključili genetske metaboličke bolesti.

Budući da se GBS nije mogao isključiti, imunoglobulin se ubrizgavao 5 dana, s ukupnom dozom od 20 g, ali dječakovi klinički simptomi nisu se značajno poboljšali. ECMO je nastavljen 4 dana dok se njegova funkcija srca i disanje nisu potpuno oporavili. Postepeno se osvijestio, ali je njegova modinamija bila toliko slaba da nije mogao micati udove, a ventilator se nije mogao odviknuti zbog slabog spontanog disanja.

Tokom hospitalizacije obavljeni su laboratorijski testovi i pregledi koji su dali sljedeće rezultate: Spektar aminokiselina i estera acilkarnitina u krvi je normalan. Tokom gladovanja, nivoi laktata u plazmi bili su 9,8 mmol/L (N: 0,5–1,7 mmol/L). EKG je otkrio periferna neurogena oštećenja višestruka demijelinizirajuća mješovita aksonska oštećenja korijena proksimalnih živaca i distalnih živčanih vlakana te oštećenja motornih i osjetnih živčanih vlakana obaju udova, donji udovi su bili gori od gornjih. Uočen je jasan spontani potencijal, koji je sugerirao da je došlo do stvarne štete. MRI glave otkrio je abnormalni intenzitet signala u lijevim bazalnim ganglijima, lijevom srednjem mozgu, četvrtoj komori, jezgri cerebelarnog zuba i vratnoj leđnoj moždini (slika 1).

Iako je pregled ukazivao na neurološku bolest, ciljano liječenje nije poboljšalo kliničke simptome dječaka tijekom jednomjesečne hospitalizacije sve dok se dijagnoza nije ispravila putem HZZ-a. CES sugerira da dijete ima nedostatak piruvat dehidrogenaze-E1 alfa, što rezultira LS. Tako je započet KD i davani su lijekovi kao što su koenzim Q10, vitamin B1 i levokarnitin. Nakon drugog 1-mjesečnog tretmana dječaku je značajno poboljšana mišićna snaga, te je mogao podići gornje udove. Međutim, mišićna snaga njegovih nogu i dalje je bila slaba.

2.3 CES (panel sa 5000 gena)

CES uključuje primjenu sekvenciranja sljedeće generacije (NGS) koja se fokusira na gene za koje je utvrđeno da su povezani s bolestima i prijavljeni u Human Mutation Database® (Su et al., 2011.). Ovaj podskup egzoma trenutno sadrži približno 5000 gena (25% egzoma) i stalno se širi. Glavna prednost CES-a je u tome što smanjuje analizu podataka, tumačenje i vrijeme obrade te povećava isplativost fokusiranjem na klinički okarakterizirane gene, primjenom trojne analize i poboljšanjem kvalitete podataka. CES rezultati za dječakove roditelje bili su normalni, ali je došlo do mutacije PDHA1 pronađen je kod pacijenta. Utvrđeno je da se mutacija NM_000284.4:c.1167_1170del nalazi u 11. egzonu PDHA1, koji se nalazio na chrX:19377758 (verzija genoma: hg19) transkript NM_000284.4 (Slika 2). Lokacija ove izbrisane mutacije ponavlja se, pa smo brisanje nazvali prema pravilima imenovanja HGVS -a. Tako je mutacija rezultirala NM_000284.4: c.1167_1170del i brisanjem osnovne CAGT. Izvršeno je Sanger sekvenciranje i potvrđeno je brisanje. Ova mutacija je u skladu s modelom genetske bolesti, a povezana bolest, sindrom uzrokovan nedostatkom E1 alfa piruvat dehidrogenaze, prema bazi podataka ClinVar, u skladu je s kliničkim manifestacijama dječaka.

Slijedi proces patogenosti varijante bolesti prema ACMG standardu (Abou Tayoun et al., 2018 Richards et al., 2015). Ocjena bolesti i patogenosti koja odgovara ovom lokusu u ClinVar bazi podataka bila je nedostatak E1-α piruvat dehidrogenaze (patogen). PVS1 je odabran jer je ovaj gen bio gen čiji je gubitak funkcije (LoF) bio poznati mehanizam bolesti. Budući da je mutirani gen kod ovog djeteta bio gen s poznatim mehanizmom bolesti, izabran je PSV1. Međutim, budući da se mjesto mutacije nalazilo na 3' -u i postojalo je više transkripata, ovaj odabir dokaza trebalo je pažljivo razmotriti. Prvi je bio da je gen imao haploinsuficijenciju. ClinGen skor haploinsuficijencije bio je 3, a ExAC pLI skor 0,992, što ukazuje da postoji dovoljno dokaza za PDHA1 gen koji ukazuje na haploinsuficijenciju (https://dosage.clinicalgenome.org/clingen_gene.cgi?sym=pdha1&subject). Odabrani transkript biološki je relevantan. Učestalost stanovništva na ovom mjestu bila je 0,00001 (maksimalna vrijednost ESP6500, 1000 g i EXAC_ALL). Učestalost populacije bliskoistočne Azije u bazi podataka gnomAD bila je 0. Stopa incidencije u normalnoj populaciji bila je vrlo niska. Osim toga, predviđanje strukture proteina i funkcije predviđanja softvera ukazale su na promjene u integritetu proteina. Gore navedeni dokazi podržavaju izbor PSV1. No, nije bilo jasno je li izmijenjena regija kritična za funkciju proteina. Nakon opsežnog razmatranja, nije bilo dovoljno dokaza za direktan odabir PVS1. Stoga je, prema stupnju snage Radne grupe za tumačenje varijacija slijeda ClinGen -a (ClinGen SVI) za PSV1, snagu PSV1 trebalo detaljno odrediti. U sljedećem koraku, egzoni zahvaćeni mutacijom nisu obogaćeni visokofrekventnim LOF varijantama u općoj populaciji (ESP6500, 1000 g i EXAC_ALL), a egzon je biološki biološki relevantan. U posljednjem koraku, mutacija je uklonila manje od 10% proteina, pa je naš nivo snage PSV1 bio PVS1_Umjeren. Prema uputama ClinGen SVI-a, nivo dokaza PVS1 je trebalo smanjiti na PVS1_Moderate (Abou Tayoun et al., 2018). Budući da je mutacija kod djeteta bila ista kao i prijavljeno povezano mjesto mutacije i promjena proteina, treba odabrati dokaz PS1. I mutacija je bila de novo, pa su izabrani dokazi PS2. Ukratko, pogledajte vodič za tumačenje mutacije gena ACMG, lokacija je zadovoljila 3 dokaza (PS1, PS2 i PVS1_Moderate), a klasifikacija je ocijenjena kao patogena (Richards et al., 2015).

2.4 Predviđanje strukture proteina

Uporedna analiza primarne strukture normalnog proteina i mutantnog proteina: Mutationtasting web stranica (http://www.mutationtaster.org/index.html) korištena je za predviđanje strukture genskih mutacija. Alat Protparam softvera ExPASY (http://www.expasy.org/tools/protparam.html) korišten je za izračunavanje glavnih fizičkih i hemijskih svojstava. TMpred softver (http: // www. Ch.embnet.org/software/TMPRED_form.html) korišten je za analizu hidrofilnosti i hidrofobnosti proteina i analizu proteinskih transmembranskih regija.

Poređenje i analiza sekundarne strukture normalnog proteina i mutantnog proteina: Softver Predictprotein (https://www.predictprotein.org/) korišten je za predviđanje sekundarne strukture normalnog proteina i mutantnog proteina kako bi se utvrdilo da li postoje strukturne promjene .

Uporedna analiza tercijarne strukture normalnog proteina i mutiranog proteina: Primjena Phyre -a (http://www.sbg.bio.ic.ac.uk/

phyre/) predviđa i upoređuje tercijarnu strukturu normalnog proteina i mutiranog proteina i grubo procjenjuje utjecaj mutacije na tercijarnu strukturu proteina.

2.5 Rezultati predviđanja strukture proteina

Rezultati predviđanja su pokazali da je mutacija uzrokovala promjenu okvira aminokiselina (p.Ser390LysfsTer33). Rezultati komparativnog predviđanja i analize strukture primarne bjelančevine: protein E1 alfa podjedinice sastavljen je od 390 aminokiselina. Nakon mutacije, 31 aminokiselina je dodata u aminokiselinsku sekvencu, molekularna struktura je promijenjena, molekulska težina je povećana, koeficijent ekstinkcije povećan, a Veliki prosjek hidropatičnosti proteina povećan (Tabela 1). Predviđeno je da će PDHA1 protein je imao dvije transmembranske regije, a ova mutacija nije bila u transmembranskoj regiji. Nakon mutacije, transmembranska spiralna struktura proteina ostala je nepromijenjena. Rezultati usporedbe sekundarne i tercijarne strukture proteina prikazani su na slikama 3 i 4. Svi rezultati sugeriraju strukturnu promjenu na kraju proteina.

Broj aminokiselina Molekularna težina Formula Ukupan broj atoma Teorijski pI Koeficijenti izumiranja Veliki prosjek hidropatičnosti Alifatski indeks Procijenjeno vrijeme poluraspada Indeks nestabilnosti
Wildtype 390 43295.63 C1899H3010N540O566S26 6041 8.35 38570 −0.312 77.59 30 č 33.06
Mutant 421 46961.79 C2059H3272N592O613S26 6562 8.59 40060 −0.376 77.65 30 č 37.72


Metode

ClinGen MM-VCEP

MM-VCEP sponzorira Američko društvo za hematologiju u partnerstvu s ClinGen-om, a opisano je na https://clinicalgenome.org/affiliation/50034/. Tim MM-VCEP-a sastoji se od 34 stručnjaka sa iskustvom u ključnim domenama i uključuje kliničke genetičare, genetske savjetnike, hematologe sa stručnom obukom iz genetike, laboratorijske i istraživačke naučnike, te stručnjake za varijante kuriranja. Dodatni naglasak stavljen je na globalno predstavljanje, sa 22 institucije koje učestvuju u 6 zemalja: Australija, Francuska, Italija, Švedska, Ujedinjeno Kraljevstvo i Sjedinjene Američke Države. MM-VCEP se redovno sastaje putem dvotjednih telekonferencija i redovno se dopisuje putem e-pošte. Odobravanje MM-VCEP-a nadgleda ClinGen i sastoji se od 4 koraka: (1) definiranje grupe/članova i opsega VCEP-a (2) razvoj pravila klasifikacije gena/bolesti (3) optimizacija pravila pomoću pilot varijanti i ( 4) MMG-VCEP odobrenje od strane ClinGen-a, implementacija pravila u ClinGen-ovom interfejsu za kuriranje varijanti i podnošenje kuriranih varijanti u ClinVar bazu podataka. Za drugi korak, članovi su bili podijeljeni u 3 podgrupe koje su se fokusirale na modifikaciju funkcionalnih/računarskih/kriterijuma spajanja (Tim F), populacijskih/fenotipskih kriterija (Tim P) i segregacijskih/aleličkih/de novo kriterija (Tim S). Svi članovi su otkrili potencijalni sukob interesa kako to zahtijeva ClinGen.

ACMG/AMP specifikacije za RUNX1

Članovi MM-VCEP-a predložili su i razgovarali o promjenama postojećih klasifikacija ACMG/AMP za RUNX1 varijante zametne linije i donijeli su odluke konsenzusa putem poziva za telekonferenciju i e-pošte. Promjene kriterija uključivale su modifikacije specifične za gen ili bolest, prilagođavanja nivoa snage, opće preporuke i određene kriterije koji su se smatrali "neprimjenjivim". Za specifikacije kriterija korištene su javno dostupne baze podataka, prediktivni softver i objavljeni podaci dobiveni iz relevantnih radova. Za BA1/BS1 RUNX1-specifične populacije MAF, proračuni su napravljeni pretpostavljajući Hardy-Weinbergovu ravnotežu pomoću nedavno objavljenog internetskog kalkulatora Whiffin/Ware. 6 Dodatni napori uključivali su identifikaciju ključnih funkcionalnih domena i mutacijskih žarišta unutar njih RUNX1, definicija informativnih funkcionalnih testova i karakterizacija fenotipskih kriterija. Također su uključene preporuke za korištenje ACMG/AMP kriterija iz ClinGen -ove SVI radne grupe. 27, -29 Preliminarne i konačne ACMG/AMP specifikacije zahtijevale su potpuni konsenzus MM -VCEP -a.

Pilot varijante

Sve pilot varijante su označene upotrebom RefSeq ID -ova NM_001754.4 i NC_000021.9 (GRCh38/hg38). Varijante koje su ClinVar-u dostavile različite kliničke laboratorije imale su prioritet za klasifikaciju. Prethodna pravila su poboljšana tumačenjem skupa od 52 RUNX1 varijante, koje su odabrane da predstavljaju spektar varijanti u RUNX1, koje pokrivaju različite vrste SNV-a, kao što su missense, besmislice, mjesta spajanja, sinonimne i intronske varijante indeli, poput dupliciranja u okviru i brisanja izvan okvira, te CNV-ovi, poput intragenih delecija. Slično, pilot varijante pokrivale su širok raspon klasifikacija u ClinVar -u, uključujući nepodudarne tvrdnje (12 BEN/LBEN, 14 VUS, 20 PATH/LPATH, 4 CONF i 2 bez ClinVar tvrdnji). Varijantna klasifikacija i primijenjena pravila pregledani su na konferencijskim pozivima radi rješavanja neslaganja i postizanja konsenzusa. Osnovne informacije o individualnim fenotipovima i segregaciji s bolešću dobijene su iz literature i ClinVar podnosilaca. Statistički pristupi za izračunavanje PS4 dostupni su u dodatnim metodama. Izvršena je dalja optimizacija pravila, a diskusija sa cijelim MM-VCEP-om je pokrenuta svaki put kada se članovi ne slažu ili iznesu zabrinutost u vezi s primjenjivošću datog pravila. Kustosi su koristili ClinGen -ovo Variant Curation Interface (https://curation.clinicalgenome.org) za procjenu i dokumentiranje primjenjivih pravila za svaku varijantu. Nakon što je MM-VCEP odobren, klasificiran je RUNX1 varijante s prilagođenim okvirom koda dokaza primijenjene na varijante dostavljene su ClinVaru i označene su šifrom dokaza sa 3 zvjezdice i zastavicom FDA za priznavanje. Prvih 52 RUNX1 varijante kuracija su sada dostupne u ClinVaru i mogu im se pristupiti na https://www.ncbi.nlm.nih.gov/clinvar/submitters/507107/.


REZULTATI

Tražili smo skor haploinsuficijencije gena na koji ne bi utjecalo koliko su pojedinačni geni dobro proučeni. Da bismo konstruirali ovaj rezultat, prvo smo razmotrili niz bioloških skupova podataka i postojeća predviđanja relevantna za ovaj cilj.

Proučite pristranost bioloških metoda i postojećih pristupa

Prvo smo procijenili pristranost studija u pristupima genske mreže za predviđanje HIS gena koje koristi Khurana et al. (4), kao i ono koje zapošljava Huang et al. (3) detaljno. Khurana et al. zaposlio je šest različitih vrsta mreža i „Multinet“, mrežu koja integrira svih šest mreža (4) (vidi sliku 1). Za svaki gen izračunali su broj veza u svakoj mreži ("stupanj" gena) i broj mreža čiji je gen bio dio. Ove biološke mreže djelomično su izgrađene na temelju niskopropusnih eksperimenata provedenih samo za gene od posebnog interesa. Međutim, poznato je da odabir 'biološki interesantnih' gena za male eksperimente utječe na to koji funkcionalni odnosi (veze) između gena će biti identificirani i na taj način uključeni u trenutne reprezentacije bioloških mreža (17). Koristili smo broj objavljenih radova povezanih s genom kao mjeru koliko su geni dobro proučeni.

Kada smo razmotrili stepen do kojeg pristrasnost studije utiče na stepen gena, uporedili smo korelaciju datog rezultata (kao što je stepen gena) sa brojem Pubmed radova po genu sa korelacijom rezultata sa ljudskim makakom. dN/dS, mjera evolucijske konzervacije za koju je poznato da je veća za HIS gene (3) (slika 1, dopunska tablica S2).

Zaista, pronašli smo značajnu korelaciju s brojem Pubmedovih radova u svakoj od mreža koje je Khurana razmatrao et al. (Slika 1). Ono što je važno, za sve osim metaboličke mreže, stupanj gena ima 1,8 do 21 puta veću korelaciju s brojem objavljenih radova nego s humanim makakom dN/dS.

Slično, broj mreža u kojima svaki gen sudjeluje 2,7 puta je jače povezan s brojem objavljenih radova nego s evolucijskim očuvanjem. Značajno je da broj mreža u kojima svaki gen sudjeluje najjači je prediktor konačnog rezultata Khurane et al. („Skor esencijalnosti“) i u snažnoj korelaciji s ovim rezultatom (Spearman ρ = 0,85, str & lt 10 −100). Posljedično, skor Essentiality je 1,6 puta snažnije koreliran s brojem Pubmed radova po genu nego s evolucijskom konzervacijom.

U alternativnom pristupu zasnovanom na mreži, Huang et al. (3) razmotrili su funkcionalnu mrežu povezivanja 'HumanNet' koja integrira interakcije protein-protein, koekspresiju gena, kocitiranje gena i druge podatke kako bi ukazali na to koliko su funkcionalno slični parovi gena. Vjerojatnost haploinsuficijencije koju je predvidio Huang et al. ('Huang HIS rezultat') u velikoj mjeri se zasnivao na blizini gena upita poznatim HIS genima u HumanNet mreži (Spearman ρ = 0,59, str < 10 −100 ). Međutim, blizina poznatih HIS gena u HumanNetu također pokazuje 2,4 puta jaču korelaciju s brojem objavljenih radova po genu nego s evolucijskim očuvanjem (Dopunska tablica S2).

Da bi se izravno pokazao utjecaj pristranosti studije na performanse rezultata za predviđanje HIS gena, potrebno je potpuno znanje o fenotipskim posljedicama poremećaja gena za sve ljudske gene. Kako takvi podaci trenutno nisu dostupni, usporedili smo performanse Essentiality skora i Huang-ove ocjene haploinsuficijencije za predviđanje skupa dobro proučenih HIS gena iz OMIM-a ('OMIM HI' Tablica 1) i skupa manje proučavanih ljudskih -to-jedan ortolozi HIS mišjih gena iz projekta Sanger Mouse (Tabela 1 'SMP Viability'). The OMIM HI genes have a significantly higher number of Pubmed papers per gene than the SMP Viability genes (Mann–Whitney str < 10 −8 Supplementary Figure S2), with greater than 2-fold difference in the median. We used MCC and the area under the ROC curve (AUC) as performance metrics, comparing scores for genes in the study set to scores of random genes matched for CDS length (see Metode). Both the Essentiality score and the Huang haploinsufficiency score have significantly higher AUC and MCC for the OMIM HI set than the SMP Viability set (Figure 2a and c, Supplementary Tables S4 and S5). The better performance of these methods on the well-studied genes is consistent with the study bias inferred above.

Gene sets used to evaluate the genome-wide haploinsufficiency score and three state-of-the-art approaches

Gene set . Description . Number of genes .
OMIM HI Online Mendelian Inheritance in Man (OMIM) haploinsufficient genes (as in ( 6)) 55
OMIM HI de novoOMIM haploinsufficient genes with de novo mutations listed in OMIM (as in ( 6)) 32
CGD AD Clinical Genomic Database (CGD) autosomal dominant disease genes 550
MGI Lethality Human genes for which the heterozygous disruption of the one-to-one orthologue in mouse causes lethality (taken from Mouse Genome Informatics (MGI) database analogous to ( 6)) 88
MGI Seizures Human genes for which the heterozygous disruption of the one-to-one orthologue in mouse causes seizures (taken from MGI analogous to ( 6)) 37
SMP Viablity Genes for which the heterozygous disruption of the one-to-one orthologue in mouse yielded significantly reduced viability by weaning (taken from the Sanger Mouse Resources Portal (SMP)) 198
SMP Viability new SMP Viability genes without MGI phenotype records prior to 10 December 2012 124
Gene set . Description . Number of genes .
OMIM HI Online Mendelian Inheritance in Man (OMIM) haploinsufficient genes (as in ( 6)) 55
OMIM HI de novoOMIM haploinsufficient genes with de novo mutations listed in OMIM (as in ( 6)) 32
CGD AD Clinical Genomic Database (CGD) autosomal dominant disease genes 550
MGI Lethality Human genes for which the heterozygous disruption of the one-to-one orthologue in mouse causes lethality (taken from Mouse Genome Informatics (MGI) database analogous to ( 6)) 88
MGI Seizures Human genes for which the heterozygous disruption of the one-to-one orthologue in mouse causes seizures (taken from MGI analogous to ( 6)) 37
SMP Viablity Genes for which the heterozygous disruption of the one-to-one orthologue in mouse yielded significantly reduced viability by weaning (taken from the Sanger Mouse Resources Portal (SMP)) 198
SMP Viability new SMP Viability genes without MGI phenotype records prior to 10 December 2012 124
Gene set . Description . Number of genes .
OMIM HI Online Mendelian Inheritance in Man (OMIM) haploinsufficient genes (as in ( 6)) 55
OMIM HI de novoOMIM haploinsufficient genes with de novo mutations listed in OMIM (as in ( 6)) 32
CGD AD Clinical Genomic Database (CGD) autosomal dominant disease genes 550
MGI Lethality Human genes for which the heterozygous disruption of the one-to-one orthologue in mouse causes lethality (taken from Mouse Genome Informatics (MGI) database analogous to ( 6)) 88
MGI Seizures Human genes for which the heterozygous disruption of the one-to-one orthologue in mouse causes seizures (taken from MGI analogous to ( 6)) 37
SMP Viablity Genes for which the heterozygous disruption of the one-to-one orthologue in mouse yielded significantly reduced viability by weaning (taken from the Sanger Mouse Resources Portal (SMP)) 198
SMP Viability new SMP Viability genes without MGI phenotype records prior to 10 December 2012 124
Gene set . Description . Number of genes .
OMIM HI Online Mendelian Inheritance in Man (OMIM) haploinsufficient genes (as in ( 6)) 55
OMIM HI de novoOMIM haploinsufficient genes with de novo mutations listed in OMIM (as in ( 6)) 32
CGD AD Clinical Genomic Database (CGD) autosomal dominant disease genes 550
MGI Lethality Human genes for which the heterozygous disruption of the one-to-one orthologue in mouse causes lethality (taken from Mouse Genome Informatics (MGI) database analogous to ( 6)) 88
MGI Seizures Human genes for which the heterozygous disruption of the one-to-one orthologue in mouse causes seizures (taken from MGI analogous to ( 6)) 37
SMP Viablity Genes for which the heterozygous disruption of the one-to-one orthologue in mouse yielded significantly reduced viability by weaning (taken from the Sanger Mouse Resources Portal (SMP)) 198
SMP Viability new SMP Viability genes without MGI phenotype records prior to 10 December 2012 124

Comparison of four gene deleteriousness scores based on known disease genes and mouse models (Table 1) as well as candidate disease genes. (a) Comparison of scores based on known disease genes and mouse models using the MCC metric. (b) Comparison of scores based on known disease genes and mouse models using the AUC metric. (c) Comparison of gene scores based on candidate disease genes using the MCC metric. (d) Comparison of gene scores based on candidate disease genes using the AUC metric. The MCC takes values between −1 and 1, with higher values indicating better performance. The AUC gives the probability that a randomly chosen gene from the set has a higher score than a randomly chosen gene from the genome (accounting for length, see Materials and Methods section). Hence possible values lie between 0 and 1, with higher values indicating better performance. Each gene set was compared to random gene sets of equal size, accounting for coding-sequence length (see Materials and Methods section). The bar plots show mean values for 100 random comparison gene sets, error bars indicate standard errors. Mann–Whitney str- i q-values for comparison of scores are listed in Supplementary Tables S4 and S5.

Comparison of four gene deleteriousness scores based on known disease genes and mouse models (Table 1) as well as candidate disease genes. (a) Comparison of scores based on known disease genes and mouse models using the MCC metric. (b) Comparison of scores based on known disease genes and mouse models using the AUC metric. (c) Comparison of gene scores based on candidate disease genes using the MCC metric. (d) Comparison of gene scores based on candidate disease genes using the AUC metric. The MCC takes values between −1 and 1, with higher values indicating better performance. The AUC gives the probability that a randomly chosen gene from the set has a higher score than a randomly chosen gene from the genome (accounting for length, see Materials and Methods section). Hence possible values lie between 0 and 1, with higher values indicating better performance. Each gene set was compared to random gene sets of equal size, accounting for coding-sequence length (see Materials and Methods section). The bar plots show mean values for 100 random comparison gene sets, error bars indicate standard errors. Mann–Whitney str- i q-values for comparison of scores are listed in Supplementary Tables S4 and S5.

Large-scale datasets without study bias

To develop haploinsufficiency predictions less affected by study bias, we wished to consider large-scale biological datasets that were obtained from genome-wide data. Following a similar ethos, Petrovski et al. ( 6) proposed to measure whether each gene is depleted in common (MAF>0.1%) non-synonymous variation based on data from over 6000 exomes. They defined the RVIS as the studentized residual when the number of common non-synonymous variants was regressed on the total number of variants in each gene. However, we found that the derivation of the RVIS induced a strong correlation between gene CDS length and the absolute size of the RVIS (Pearson r = 0.5, str < 10 −100 see Metode). Moreover, both the highest and lowest RVIS values were preferentially attained by the longest genes (Supplementary Figure S3a). This is largely due to the construction of the score, as we found similar dependence on CDS when randomizing the proportion of common non-synonymous variants among genes (see Supplementary Data, Supplementary Figure S4). Hence, the effects of CDS and intolerance to gene disruptions on the RVIS are difficult to disentangle. Moreover, the RVIS does not account for potential differences in the relative numbers of possible synonymous and non-synonymous mutations in genes.

Consequently, we instead derived an alternative score for the relative depletion of common functional variation in each gene: the ratio |$frac<><>$| (vidi Metode), which we call the ‘Non-synonymous Variation Depletion Score’ or NoVaDs. Intuitively, the intolerance of a population to functional variants in a given gene will act to decrease the MAF of such variants, thus decreasing the NoVaDs. The NoVaDs is not correlated with the CDS length of genes (r = −0.04, str < 10 −5 Supplementary Figure S3b). Importantly, when gene length is accounted for, the NoVaDs distinguishes disease genes from random human genes better than the RVIS (Supplementary Figure S3c and d, Supplementary Table S6).

As expected, the NoVaDs is 1.36-fold more highly correlated with evolutionary conservation (Spearman ρ = 0.35, str < 10 −100 ) than with Pubmed papers per gene (ρ = −0.26, str <10 −100 ), thus not showing study bias.

As the correlation between NoVaDs and evolutionary conservation is only moderate, we also applied the NoVaDs to evaluate the biological networks considered by Khurana et al. ( 4) and Huang et al. ( 3) for study bias (Figure 1, Supplementary Table S2). The results showed evidence for study bias entirely consistent with the observations based on evolutionary conservation, hence providing additional evidence for bias in those networks.

By contrast, we found no evidence for study bias in large-scale gene co-expression networks (Figure 1, Supplementary Table S2), specifically COEXPRESdb (based on microarray data ( 18)) and a network constructed from the pilot 1 phase RNA-sequencing data of the Gene-Tissue Expression Consortium (GTEx ( 19)). Similarly to the approach of Huang et al. ( 3), we considered how strongly each gene is co-expressed with 297 known HIS genes ( 7) (see Metode). This value is strongly correlated with the gene degree in the networks (COEXPRESdb: ρ = 0.92 GTEx: ρ = 0.77 both str < 10 −10 ), and does not show a study bias (Supplementary Table S2).

Unbiased haploinsufficiency score predictions

To derive a score indicating how likely each human gene is to be HIS, we applied a machine learning method (an SVM see Metode) to a range of gene features. Based on the results above, we used gene features that did not show study bias for the predictions, namely the co-expression with known HIS genes in the COEXPRESdb and GTEx co-expression networks the NoVaDs evolutionary conservation and the ratio of gene expression in fetal to adult tissue (see Metode). As with other methods, to train the SVM we used HIS genes taken from a review ( 7) while HS genes were obtained as genes disrupted by deletion copy number variants in healthy individuals ( 8) (see Metode). HS genes were subsampled 100 times, averaging predicted HIS scores for each gene (see Metode). As our method is applicable to genes irrespective of their degree of study, we called the resulting score ‘Genome-wide haploinsufficiency score’ or GHIS.

We wanted to compare the GHIS to previously published methods, denoting the latter scores as ‘Essentiality score’ ( 4), RVIS ( 6) and ‘Huang HIS score’ ( 3) (see Metode). The GHIS provides a score for a higher number of genes than previously published methods (Table 2). Crucially, of the genes with a provided score, the GHIS includes about twice as many genes currently unassociated with any Pubmed papers as each of the three previously published methods, both in absolute numbers and as a proportion of the total predictions (Table 2).

The genome-wide haploinsufficiency score evaluates a higher number of genes than three previously published methods, as well as a higher number of less-studied genes

Score . Ukupno . Without Pubmed paper . Percent without Pubmed paper .
GHIS 19 701 4621 23.46%
Huang HIS score 17 069 2064 12.09%
Essentiality score 18 386 1525 8.29%
RVIS 16 572 1774 10.70%
Score . Ukupno . Without Pubmed paper . Percent without Pubmed paper .
GHIS 19 701 4621 23.46%
Huang HIS score 17 069 2064 12.09%
Essentiality score 18 386 1525 8.29%
RVIS 16 572 1774 10.70%

Essentiality score = Khurana et al. ( 4) gene indispensability score RVIS = Petrovski et al. ( 6), Residual Variance Intolerance Score Huang HIS score = Huang et al. ( 3) haploinsufficiency probabilities.

Score . Ukupno . Without Pubmed paper . Percent without Pubmed paper .
GHIS 19 701 4621 23.46%
Huang HIS score 17 069 2064 12.09%
Essentiality score 18 386 1525 8.29%
RVIS 16 572 1774 10.70%
Score . Ukupno . Without Pubmed paper . Percent without Pubmed paper .
GHIS 19 701 4621 23.46%
Huang HIS score 17 069 2064 12.09%
Essentiality score 18 386 1525 8.29%
RVIS 16 572 1774 10.70%

Essentiality score = Khurana et al. ( 4) gene indispensability score RVIS = Petrovski et al. ( 6), Residual Variance Intolerance Score Huang HIS score = Huang et al. ( 3) haploinsufficiency probabilities.

Predictions for genes with disease association

In the next step, we evaluated the different scores on gene sets with known disease association (Table 1).

These genes have higher CDS length than general human genes (Supplementary Figure S5a). Consequently, we compared each gene set to 100 random gene sets with the same number of genes, matching genes for CDS length (see Metode). As in previous studies, MCC was used as primary comparison metric, assessing how many disease genes versus random genes fell into the genes predicted to be among the 25% most intolerant to disruption (cut-off chosen as used by Petrovski et al. ( 6) see Metode). We compared scores with the Mann-Whitney rank test.

The GHIS performed as well as or significantly better than the RVIS on all gene sets, and at least as well as the Essentiality score on all but the ‘OMIM HI’ and ‘MGI Lethality’ gene sets (all q < 10 −9 Figure 2a, Supplementary Table S4).

The Huang HIS score outperformed the GHIS on the ‘OMIM HI’, ‘OMIM HI de novo’, ‘CGD AD’ and ‘MGI Lethality’ gene sets (all q < 10 −10 ). However, these are some of the most studied human genes (median >50 papers/gene Supplementary Figure S2a). By contrast, performance of the Huang HIS score declined steadily for less-studied genes, and the GHIS performed better than all published methods when predicting the considerably less-studied ‘SMP Viability’ and ‘SMP Viability new’ genes, as well as the ‘MGI Seizure’ genes (all q < 10 −10 ).

The results were similar when considering the area under the ROC curve (AUC see Figure 2b, Supplementary Table S5), although with better relative performance of the RVIS and Essentiality score. Notably, the GHIS again significantly outperformed all of the three other scores on the ‘SMP Viability’ and ‘SMP Viability new’ genes, as well as the ‘MGI Seizure’ genes (all q < 10 −10 ).

Predictions for disease candidate genes

Finally, we considered these methods’ predictions made for sets of disease candidate genes from recent exome sequencing studies. Autism probands have an elevated rate of de novo LoF mutations than unaffected individuals ( 20) around half of the de novo LoF mutations are expected to be causal ( 20), suggesting that the corresponding genes are HIS. As only about 50% of the ASD genes are likely to be causal, even with a score that distinguishes perfectly between HIS and HS genes, the MCC and AUC are expected to be lower than 1 for the ASD genes. Under the best-case scenario, we would expect the MCC to be around 0.38 and the AUC around 0.75 (see Supplementary Data).

We considered two independent sets of de novo LoF genes in autism (‘ASD1’, n = 50 ( 12) ‘ASD2’, n = 49 ( 13–15)) and their combination (‘ASD12’, n = 98). Genes with de novo mutations tend to have high CDS length (Supplementary Figure S5b), hence we accounted for CDS as for the disease gene sets above. As expected, the MCC and AUC for all four scores considered in this study are lower than under the best-case scenario (Figure 2c,d).

However, the GHIS significantly outperformed all three previously published scores on all three autism gene sets using the MCC metric (Mann–Whitney q < 10 −6 for all three comparisons, Figure 2c, Supplementary Table S4), as well as using the AUC metric (Mann–Whitney q < 10 −10 for all three comparisons Figure 2d, Supplementary Table S5).

For the genes in the ASD1, ASD2 and ASD12 sets, we do not know which are causal, and therefore cannot evaluate the accuracy of the methods further. Consequently, as another example of practical application, we considered 18 genes disrupted by de novo LoF mutations in at least two autism probands from a larger study ( 16) (‘ASD_M’). All of these genes are associated with autism at <10% FDR based on de novo and transmitted genetic variants ( 16). Therefore, these genes should rank highly on a HIS score. We asked how many of the 18 ASD_M genes fell into the genes with the top 1, 2, …, 99% score for the GHIS and the three previously published scores. These genes have high CDS lengths (Supplementary Figure S6a), and their CDS is strongly correlated with their RVIS (Spearman ρ > 0.8) in agreement with the CDS length bias described above. Consequently, the results for the RVIS and the CDS are extremely similar (Supplementary Figure S6a) and it is difficult to quantify to which extend the ranking of these genes is confounded by mutations being more frequent in longer genes. While we accounted for the CDS-bias in the above analyses through randomizations, for this straightforward application, no like-for-like comparison of the RVIS to the three other methods was possible.

Of the remaining three scores, the GHIS performs at least as well as the Huang HIS score and the Essentiality Score across all possible cut-offs (Supplementary Figure S6b).

These results suggest that none of the methods considered in this study are accurate enough for use in a clinical setting, but that the GHIS has relatively the best performance on the best autism candidate genes.


FUTURE PLANS

The coming year will bring more data and tools to the UCSC Genome Browser. Additional features will be added to the phased haplotypes display. More track filtering options will be added, and the ‘hide empty subtracks’ feature will include a display of the number of tracks that are hidden in the viewing window. New features for composite tracks are in development such as introducing faceted search controls to configure complex composite tracks. Support for single-cell sequencing will continue to be developed in the coming year. We will continue to incorporate COVID-19 human annotations and SARS-CoV-2 viral genome annotations as they become available.


Baza rijetkih bolesti

NORD gratefully acknowledges Eduardo Pérez Palma, PhD, Cologne Center for Genomics, University of Cologne, Germany Dennis Lal, PhD, Genomic Medicine Institute, Lerner Research Institute, Cleveland Clinic, Cleveland, USA Katrine M. Johannesen, MD, The Danish Epilepsy Center Filadelfia, Dianalund, Denmark, and SLC6A1 Connect for the preparation of this report.

Synonyms of SLC6A1 Epileptic Encephalopathy

  • SLC6A1-related myoclonic-atonic epilepsy (MAE)
  • SLC6A1 haploinsufficiency / loss of function
  • GAT1 deficiency
  • SLC6A1-related disorders

Opšta diskusija

SLC6A1 epileptic encephalopathy is an autosomal dominant genetic disorder characterized by the loss-of-function of one copy of the human SLC6A1 gen. Clinical manifestation of SLC6A1 epileptic encephalopathy is characterized by early onset seizures (mean onset 3.7 years) and mild to severe intellectual disability. Seizure types include absences, myoclonic and atonic seizures. Language impairment and behavioral problems have also been observed. Some patients have shown intellectual disability without seizures or associated with focal epilepsy.

Znakovi i pojačani simptomi

Given the limited number of patients available for characterization, the full extent of symptoms is yet to be described. The most common features observed include absence seizures, myoclonic-atonic epilepsy (onset from 7 months to 6 years, mean 3.7 years) and mild-to-moderate intellectual disability. Speech difficulties and behavioral problems have been described. The most common EEG pattern observed comprises irregular, high ample, and generalized spike-and-waves. To date, the most extensive patient collection was published by Johannesen et al5 and includes 34 patients. In this cohort, cognitive development was impaired in 33/34 (97%) subjects 28/34 (82%) had mild to moderate intellectual disability, with language impairment being the most common feature. Epilepsy was diagnosed in 31/34 patients with a mean onset at 3.7 years. Cognitive assessment before epilepsy onset was available in 24/31 subjects and was normal in 25% (6/24). Two patients had speech delay only, and 1 had severe intellectual disability. After epilepsy onset, cognition declined in 46% (11 out of 24) of patients. The most common seizure types were absences, myoclonic, and atonic seizures. Sixteen patients (47%) fulfilled the diagnostic criteria for myoclonic-atonic epilepsy. Seven additional patients had different forms of generalized epilepsy, and two had focal epilepsy. Electroencephalography (EEG) findings were available in 27/31 patients showing irregular bursts of diffuse 2.5-3.5 Hz spikes/polyspikes-and-slow waves in 25/31. Two patients developed an EEG pattern resembling electrical status epilepticus during sleep. Ataxia was observed in 7 out of 34 patients (21%).

Uzroci

SLC6A1 epileptic encephalopathy is caused by a change (mutation) in one copy of the SLC6A1 gene that prevents the gene from working properly.

Two types of SLC6A1 gene variants have been observed in patients: 1) protein truncating variants that stop the protein production for one of the two SLC6A1 genes inherited from parents and 2) point mutations in critical regions of the protein such as GABA binding sites and transmembrane domains, which lead to loss-of-function of mutated proteins. Thus, the expected molecular pathomechanism of SLC6A1 disorders is haploinsufficiency a single functional copy of the gene is not enough. The disease-mode is supported by experiments in GAT-1 knockout mice as well as mice administered with GAT-1 inhibitor. In these experiments, the mice show spontaneous spike-wave discharges typical of absence seizures, the predominant seizure type seen in individuals with SLC6A1 mutacije. Recently, experimental evidence showed that SLC6A1 variants identified in epilepsy patients reduce GABA transport6 in vitro.

The SLC6A1 gene encodes for the voltage-dependent c-aminobutyric acid (GABA) transporter 1 (GAT-1) protein, one of the major GABA transporters of the human central nervous system. SLC6A1 is primarily expressed in the brain, specifically, in GABAergic neurons and astrocytes. Primarna funkcija od SLC6A1 is the reuptake of the GABA neurotransmitter from the extracellular space in the synapsys1. The SLC6A1 gene is located in the short arm of chromosome 3 (GRCh38 genomic coordinates: 3:10,992,733-11,039,248), contains 15 exons and is approximately 25 kb long. Genetic variation affecting the coding sequence of the gene in the general population is extremely rare2. Dakle, SLC6A1 gene is highly intolerant to variation. Patient variants in SLC6A1 were first described by Carvill et al. in 20153 (Online Mendelian Inheritance in Man database (OMIM) 137165)4 and were associated with early onset myoclonic-atonic epilepsy and intellectual disability. Later, with the collection of more patients, the phenotypic spectrum of SLC6A1-related disorders was expanded to include several types of seizures and different degrees of developmental delay. Notably, almost all of the genetic variants reported to date were not present in the parents (they arose ‘de novo’) and have not been observed in the general population.

SLC6A1 epileptic encephalopathy is an autosomal dominant genetic condition. Dominant genetic disorders occur when only a single copy of a non-working gene is necessary to cause a particular disease. The non-working gene can be inherited from either parent or can be the result of a mutated (changed) gene in the affected individual. The risk of passing the non-working gene from an affected parent to an offspring is 50% for each pregnancy. The risk is the same for males and females.

Pogođene populacije

This is an extremely rare disorder. To date, only 34 patients have been characterized in the literature5. Patients are from families with various ethnic backgrounds from the USA, Canada and European countries. The SLC6A1 gene was until recently not screened in diagnostic sequencing and it is likely that many more patients will be reported with inclusion of this gene on gene panels.

Srodni poremećaji

A large number of infantile epileptic encephalopathies are known which result from mutations in various genes (including KCNQ2, FOXG1, etc.) and show overlapping characteristics (phenotypes) with SLC6A1 nedostatak. Specifically, GABA transporters are part of the large family of neurotransmitters known as sodium symporters and include 13 gene members that have highly conserved sequence and redundant functions. For five additional gene family members, disease-causing mutations have been described in public repositories7: SLC6A2, SLC6A3, SLC6A5, SLC6A8 i SLC6A9 which are associated with a broad spectrum of neurodevelopmental disorders including epilepsy and intellectual disability. Genetic variants in the genes SCN1A, SCN1B, GABRG2, i CHD2 can cause similar phenotypes including myoclonic atonic seizures and response to valproic acid.

Dijagnoza

In order to diagnose a SLC6A1 epileptic encephalopathy, DNA sequencing is required. Depending on the available resources, whole genome and whole exome sequencing can be performed. However, targeted gene panel sequencing is often faster, less expensive and easier to reimburse by insurance. The SLC6A1 gene is included in a variety of current epilepsy-oriented gene panels. Independently of the sequencing method used, variants found in the SLC6A1 genes should be interpreted carefully. The American College of Medical Genetics (ACMG) guidelines should be followed to assign the variants found a disease-causing state8.

Standardne terapije

Liječenje
Currently, the number of patients and clinical data available is limited. Treatment may require interdisciplinary efforts including neurologists, developmental pediatricians, speech therapy and/or other health care professionals to systematically and comprehensively plan the treatment of an affected child. Data on drug treatment is not conclusive. However, valproic acid by itself or in combination with other antiepileptic drugs has shown positive results. Johannesen et al5 shows that ten out of 15 patients treated with valproic acid became seizure-free, and the remaining 5 showed a partial benefit. Valproic acid is thought to have a positive effect on the GABA system, possibly by increasing the GABA concentration in the human brain9. Overall, 20 out of 31 patients became seizure-free, with valproic acid being the most effective drug. There was no clear-cut correlation between seizure control and cognitive outcome.

Istražne terapije

SLC6A1 Connect is partnering with Dr. Steven Gray from UT Southwestern to develop a gene replacement therapy to treat SLC6A1 mutations. Pre-clinical and experimental work is currently underway aiming to produce a custom adeno-associated virus (AAV) suitable for SLC6A1 treatment.

Information on current clinical trials is posted on the Internet at https://clinicaltrials.gov/. Sve studije koje su finansirane od strane američke vlade, a neke su podržane od strane privatne industrije, objavljene su na ovoj vladinoj web stranici.

Za informacije o kliničkim ispitivanjima koja se provode u Kliničkom centru NIH u Bethesdi, MD, kontaktirajte Ured za zapošljavanje pacijenata NIH:

Za informacije o kliničkim ispitivanjima sponzoriranim od privatnih izvora, kontaktirajte:
http://www.centerwatch.com/

Za informacije o kliničkim ispitivanjima provedenim u Europi obratite se:
https://www.clinicaltrialsregister.eu/

Organizacije članice NORD-a

Druge organizacije

    • poštanski fah 8126
    • Gaithersburg, MD 20898-8126
    • Telefon: (301) 251-4925
    • Besplatni telefon: (888) 205-2311
    • Web stranica: http://rarediseases.info.nih.gov/GARD/

    Reference

    1. Scimemi, A. Structure, function, and plasticity of GABA transporters. Frontiers in Cellular Neuroscience 2014 8: 161.

    2. Lek M, Karczewski KJ, Minikel EV, et al. Analysis of protein-coding genetic variation in 60,706 humans. Nature 2016536: 285-291.

    3. Carvill GL, McMahon JM, Schneider A, et al. Mutations in the GABA transporter SLC6A1 cause epilepsy with myoclonic-atonic seizures. AJHG 2015 96: 808-815.

    4. Amberger JS, Bocchini CA, Schiettecatte F, Scott AF, and Hamosh A. OMIM.org: Online Mendelian Inheritance in Man (OMIM), an online catalog of human genes and genetic disorders. Nucleic Acids Research 2015 43: D789-798.

    5. Johannesen KM, Gardella E, Linnankivi T, et al. Defining the phenotypic spectrum of SLC6A1 mutations. Epilepsia 201859: 389-402.

    6. Mattison KA, Butler KM, Inglis GAS, et al. SLC6A1 variants identified in epilepsy patients reduce gamma-aminobutyric acid transport. Epilepsia 201859: e135-e141.

    7. Landrum MJ, Lee JM, Benson M, et al. ClinVar: public archive of interpretations of clinically relevant variants. Nucleic Acids Research 2016 44, D862-868.

    8. Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genetics in Medicine 201517: 405-424.

    9. Chateauvieux S, Morceau F, Dicato M, and Diederich M. Molecular and therapeutic potential and toxicity of valproic acid. Journal of Biomedicine & Biotechnology 2010 Published online Jul 29.

    Godina objavljivanja

    Podaci u bazi podataka rijetkih bolesti NORD & rsquos su samo u obrazovne svrhe i nemaju namjeru zamijeniti savjete ljekara ili drugog kvalifikovanog medicinskog stručnjaka.

    Sadržaj web stranice i baze podataka Nacionalne organizacije za rijetke poremećaje (NORD) zaštićen je autorskim pravima i ne smije se reproducirati, kopirati, preuzimati ili distribuirati, na bilo koji način, u bilo koju komercijalnu ili javnu svrhu, bez prethodnog pismenog odobrenja i odobrenja NORD-a. . Pojedinci mogu odštampati jedan štampani primjerak pojedinačne bolesti za ličnu upotrebu, pod uvjetom da je sadržaj neizmijenjen i uključuje autorska prava NORD & rsquos.

    Nacionalna organizacija za rijetke poremećaje (NORD)
    55 Kenosia Ave., Danbury CT 06810 & bull (203)744-0100


    View Variant Details

    View more information about a variant, including gene, transcript, and association details. TruSight Software provides access to curated data from public knowledge bases, and a private knowledge base of associations created by members of your workgroup.

    To view variant details, select the row containing the variant.

    Lists information about a variant. Data vary depending on variant type.

    The chromosome number and position.

    Short tandem repeat details.

    Low Complexity Region (gnomAD)

    The variant is found in a low-complexity region (LCR).

    Links to visualization browsers.

    Links to public database entries for the variant. ClinVar entries display the number of stars associated with review status.

    Lists the zygosity and variant read metrics for each subject.

    The affected status of the subject.

    The genotype of the variant.

    Variant Read Frequency (VRF)

    The fraction of reads at this position that have the variant allele.

    The total number of reads passing quality filters at this position.

    The number of split reads.

    The number of paired reads.

    The number of reads called for the ALT Allele.

    The status of the variant call quality as annotated in the VCF file. PASS indicates that all filters were passed. Otherwise, the variant call filter is listed.
    The filter listed and threshold for passing filter depends on the method of generating the VCF file.

    The numeric value of variant call quality. Determination of variant quality depends on the variant caller.

    Lists the CaseLog data for the subject.

    Alternative Allele Frequency

    The frequency of the selected alternate allele within your aggregated cases in CaseLog.

    The number of cases, including cases from public and shared data, in which a variant was observed or reported.

    Selecting More Info displays details about the cases.

    Lists available MITOMAP data for small variants.

    The associated phenotypes in the MITOMAP database

    The presence of heteroplasmy (mixture of mutant and normal mtDNAs)

    The presence of homoplasmy (pure mutant mtDNAs).

    The status of pathogenicity reports. Possible values are:

    The significance of the variant.

    The number of GenBank full-length sequences.

    Links, if available, to PubMed entries for the variant.

    Lists the maximum population frequency in each data source. The population databases vary depending on the variant type.

    Selecting All Population displays detailed population data.

    For information about data sources and population groups, see Population Data.

    Lists available SpliceAI predictive scores for splice junctions: acceptor gain, donor gain, acceptor loss, or donor loss. For more information, see https://github.com/Illumina/SpliceAI.

    The probability that the variant affects the category. The maximum probability across all categories is reported in the variant grid.

    The distance between the splice change relative to the variant position. Positive values are downstream of the variant, negative values are upstream.

    Lists the results of the SMN determination of whether wildtype SMN1 is absent. If the gene is absent, the result is reported as Detected.

    Lists the associations between the variant and disease, and Association Details za svaki.

    Indicates whether the association is selected for inclusion in the report.

    The predicted interpretation for a disease. The color of the text indicates the interpreted pathogenicity.

    The date the association was added.

    The version number of the association.

    The association level. Possible values are:

    Nucleotide —Specific alterations frequently observed in the disease condition or mentioned in the evidence.

    Other association levels are not supported in this release.

    The description of the association provided by the curator.

    A link to the Knowledge Network entry for the association.

    Lists comments for the variant. Za više informacija pogledajte Manage Comments.

    Do not include direct identifiers in comment fields.

    Lists details about the variant allele and consequence.

    The transcript with the longest coding sequence or the longest complementary DNA.

    The name of the selected transcript, typically a database identifier from RefSeq or Ensembl, linked to the transcript page in the database. TruSight Software includes annotations for all transcripts that overlap a variant.

    The names of genes associated with the transcript.

    The probability that the effect of an amino acid substitution on the function of a human protein is tolerated.

    The predicted effect based on SIFT score.

    The probability that the effect of an amino acid substitution on the function of a human protein is damaging.

    The predicted effect based on PolyPhen score.

    The REVEL predicted pathogenicity of a missense variant.

    The N-species amino acid conservation score.

    The HGVS coding nomenclature.

    The HGVS protein nomenclature.

    The consequence of the variant, described in Sequence Ontology standardized vocabulary.

    Lists details about the affected genes.

    If more than one gene is affected, the initial view lists a summary of each gene and any overlapping phenotypes. A partial overlap icon indicates that a breakend occurs within the gene.

    Select a gene name to view details.

    The gene name and summary description.

    Constraint Metrics (gnomAD)

    Scores for synZ, misZ, pNull, pLI, pRec, and LOEUF metrics.

    Dosage Sensitivity (ClinGen)

    The ClinGen evidence classification for haploinsufficiency and triplosensitivity.

    The associated phenotypes in the ClinGen database. Colored icons represent the ClinGen classification of the disease.

    The predicted pathogenicity of missense variants in primate species. Values closer to 1 indicate greater pathogenicity.

    For more information about the PrimateAI prediction engine, see https://github.com/Illumina/PrimateAI.

    The results of a literature search for the association, linked to the search results. The label indicates whether overlapping indications exist, followed by the number of results for the gene.

    Links, if available, to database entries for the gene (eg, OMIM, PubMed, or CaseLog).

    Case-independent comments about the gene. Za više informacija pogledajte Manage Comments.

    Do not include direct identifiers in comment fields.

    Lists the associations between the gene and disease, and Association Details za svaki.

    Indicates whether the association is selected for inclusion in the report.

    The predicted interpretation for a disease. The color of the text indicates the interpreted pathogenicity.

    The date the association was added.

    The version number of the association.

    A link to the Knowledge Network entry for the association.

    Summary of variants that have been observed for a gene in other cases.

    Selecting Full view in CaseLog displays the CaseLog page with details about the gene and variant. For more information about CaseLog data, see CaseLog.

    Lists flag and review activity in the case.

    Activities do not immediately update in the log.

    Innovative technologies

    At Illumina, our goal is to apply innovative technologies to the analysis of genetic variation and function, making studies possible that were not even imaginable just a few years ago. It is mission critical for us to deliver innovative, flexible, and scalable solutions to meet the needs of our customers. As a global company that places high value on collaborative interactions, rapid delivery of solutions, and providing the highest level of quality, we strive to meet this challenge. Illumina innovative sequencing and array technologies are fueling groundbreaking advancements in life science research, translational and consumer genomics, and molecular diagnostics.


    Pogledajte video: Webinar: NCBI Resources and Variant Interpretation Tools for the Clinical Community (Oktobar 2022).