Analysis of 200,000 exome-sequenced UK Biobank subjects illustrates the contribution of rare genetic variants to hyperlipidaemia

preprint OA: closed
📄 Open PDF Full text JSON View at publisher
AI-generated summary by claude@2026-07, 2026-07-16

Rare genetic variants in genes like LDLR and PCSK9 significantly contribute to hyperlipidaemia risk in a large UK Biobank cohort, with deleterious variants present in about 1.5% of the population.

One-sentence paraphrase of the abstract; not a substitute for reading it. No clinical advice. How this works

AI-generated deep summary by claude@2026-07, 2026-07-16 · read from full text

This study used weighted burden analyses on exome-sequenced data from 200,000 UK Biobank participants with hyperlipidaemia (44,050 cases and 156,578 controls) to assess whether rare coding variants contribute to hyperlipidaemia risk, using signed log10 p-values (SLP) as the association metric. After accounting for ancestry principal components and sex (to address a tendency toward more rare X-chromosome variants in females), the analyses strongly implicated LDLR (SLP = 50.08) and PCSK9 (SLP = -10.42), while also highlighting additional genes previously associated with lipid levels. The paper reports that SIFT-classified deleterious variants show an average two-fold effect and cumulative frequency consistent with variants present in approximately 1.5% of the population, emphasizing that large exome datasets can detect many moderately acting protein-altering variants. It notes limitations around variant functional prediction and that optimal weighting schemes for burden analyses have had limited empirical evidence. The paper does not explicitly discuss endometriosis or adenomyosis; it was included in the corpus via a keyword match in the upstream search index.

Read from the paper's body, not the abstract. Not a substitute for reading the paper. No clinical advice. How this works

Abstract

A few genes have previously been identified in which very rare variants can have major effects on lipid levels. Weighted burden analysis of rare variants was applied to exome sequenced UK Biobank subjects with hyperlipidaemia as the phenotype, of whom 44,050 were designated cases and 156,578 controls, with the strength of association characterised by the signed log 10 p value (SLP). With principal components included as covariates there was a tendency for genes on the X chromosome to produce strongly negative SLPs, and this was found to be due to the fact that rare X chromosome variants were identified less frequently in males than females. The test performed well when both principal components and sex were included as covariates and strongly implicated LDLR (SLP = 50.08) and PCSK9 (SLP = -10.42) while also highlighting other genes previously found to be associated with lipid levels. Variants classified by SIFT as deleterious have on average a two-fold effect and their cumulative frequency is such that they are present in approximately 1.5% of the population. These analyses shed further light on the way that genetic variation contributes to risk of hyperlipidaemia and in particular that there are very many protein-altering variants which have on average moderate effects and whose effects can be detected when large samples of exome-sequenced subjects are available. This research has been conducted using the UK Biobank Resource.
Full text 79,019 characters · extracted from oa-pdf · 5 sections · click to expand

Introduction

We rec e ntly rep ort ed th e re sul t s of an aly s i s of 5 0,000 ex ome- s equenc e d UK Bioba n k subjec ts ai ming to ide n t ify ra re va r ia n t e ff ect s in ge ne s in fluenc ing su scep tibil ity to hy pe rlipidae m ia an d al so brie fly revie wed w hat wa s k nown t o da te ab out the gene tic contrib uto rs t o t hi s ph e noty pe (C urti s, 2020 ). The poten tial a dvan tag e o f st u dying r a r e variant s is tha t they ha ve mo re pr o fou n d, readily interp re table impac ts on bi ology t h an co m mon va r ia nts, w ho s e e f fect siz es ten d to be c onstrai ned by s el ec tion p res su r e s. R a r e vari ant s wit h a large dominan t e ff ec t in LD L R , APO B and PC S K 9 c ause 40% of ca ses o f fam ilial hyperli pi daemia and the re a re al so commo n varia n ts wh ich e xert s m a ll ef fec ts on h yp erlipida emia r i sk (N ata raja n et al. , 2018; S hari fi e t a l., 2017 ; Willer e t al ., 2013 ; Wu et al. , 2019) . Al t h ou gh fo r mos t gene s impa i r e d func tion incr ea se s ri s k , t he PCS K9 va ria nt s w hich caus e famil ial hy perl ipida emi a produc e a gain of fu nction w her ea s lo s s o f f u nc tion v ari ant s cau se All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice. hypo betalip opro tein e mia an d PCSK9 inhi bitor s ar e u sed a s t r e a tmen ts to lower c h ole st e r ol leve l s (Dron and Hege l e, 20 16) . The pr e viou s ana ly si s of 50,000 U K Biob a nk identifi ed on e ge ne, HUWE1, whic h met c r i te r ia for sta t i s t ic al signi fic anc e a f ter cor rec t i on fo r multiple t e sting a nd in w hich t h e r e wa s an exc ess o f rar e and/o r dama ging variant s in c on tr ol s , su gg es ting t hat impai r ed functioni ng of thi s gene wa s protec tiv e aga in st h ype rlipida emia. A nu mber of o the r ge ne s whic h were individ u ally s ig nifica n t with unc or rec ted p<0 .0 01, w ere arguabl y o f poten tia l int ere s t , includ ing LD LR , an d , in an analy s i s o f se ts of ge ne s , t h e GO g en e s et GEN E RATI O N OF PRE C U RS OR MET ABO L ITES AND ENER GY wa s sta ti st i call y signific ant . The whole U K Biobank s a mpl e c ons i st s of 500,00 0 s u bject s an d a new r e lea s e of d a t a mea ns tha t the re i s no w ex ome s eque nc e data avai l able f o r 2 00,000 o f t h em (Szu s t ak owski e t a l., 2020 ). We repo rt he re th e re sul t s of a nal ysis of t his l arger d a t a set, w hich inclu de s the original 50,0 00. La r g e sampl e s of ex ome s equen ce d su bjec t s ha ve onl y bec ome a vail a ble re la t i vel y r e cently and co nt rov ersy rema in s abou t the o ptima l method s o f ana ly si s . Seq ue ncing r e ve al s v ery large number s of ge ne tic v arian ts , many of whic h wi ll hav e no b iologic al e ffe ct and /or wil l be e x tr e mely r a r e, oc curring in only a handful of s ubj ect s o r jus t a s singl e ton s . The r arity o f i ndividu al va r i ant s me an s that they nee d to be gro uped t oge t h e r in a bu r de n anal y s i s a nd i t is comm on pr actic e to c ombin e a ll va r i ant s w hich are pr edic te d to comp l ete ly di s rupt the w or k ing o f a gen e, com p r is ing : varian ts wh ich introduc e a s top codo n ; s m a ll ins er tion s and del e tion s whic h are no t a multipl e of t h r ee b a se s and henc e di srup t the amin o aci d code , term ed fr ame shif t varia n ts; va r i an t s c hang ing es se ntial splic e si te seq uenc e s a t intro n-e xon bo unda rie s, di s r up t in g normal splic ing o f exon s. Th e s e three typ e s of va r i ant ar e pre dic ted to all have a br oadl y similar e ff ec t no ma tte r whe r e t h ey oc c ur in the gen e, co ns i sti ng of a com ple te fail ure of the ge ne to pro duce normal p roduc t, and th ey may be ref erre d to as lo s s o f func tion (L OF ) varian t s. I t may t hen bec ome po s sibl e to i mplica t e a ge n e in the pathog en e si s o f ph eno type by obs erving a ge neral ex ces s o f LOF v arian t s in tha t ge ne among ca s e s relativ e to co ntrol s (Sin gh e t al . , 2016). How ever it i s c erta inly t he ca s e tha t oth er k inds o f varia nt ca n also c au s e di sea s e. A v aria nt whi ch c hange s a codon s o tha t i t cod e s fo r a dif f eren t a mino ac id, termed a non -s y nonymou s va r i a nt, may alter the st r uc tu r e or functio n of the pro t e in pro duct in a wa y wh ich dram atical ly a f f e ct s ri sk bu t al tern ativel y a p r o tein al ter ing v ariant may hav e no ef fec t at all . The impac t o f a non - synony mou s v ari ant w ill dep end c r uc i ally on the n a ture of the ami no ac id ch ange and it s po s i ti on in th e p r ot ein an d it rema in s a cha ll eng ing ta sk to pr edi ct the biol ogic al ef fec t a lthough c ommonly u sed so ft w ar e such a s Pol y Phen and S I FT at tempt thi s ( Adz h ubei et a l ., 2013 ; Kumar et a l. , 2009 ). PolyP h en de si gna t e s some v arian t s a s “pos s i bly da mag ing ” or “ pr o bably dama ging” and SIFT designa te s some va ri ant s a s “ dele terio us” bu t dif fer ent p redi ction pr o gram s do not al way s a g r e e wi t h each oth er. N on sy no nymous varia n ts ar e much more frequ e nt th an LOF va r i ant s a nd so it woul d b e de sir able to incorpora te them into bu rden a nalys e s b ut the re i s a ri sk that doi ng so may simpl y introduc e addit iona l noi se . Ide n t i fying whic h s p e cific va ria nts ar e most lik ely t o have biologi cal e ffe ct s could incr ea se po wer to impli cat e ri sk g ene s but re mains a ch alleng ing ta sk. Even v arian t s whic h do no t c hang e amin o aci d s e quence , inc ludi ng syno nymous and intr o ni c v ariant s , can t hroug h vario u s mec hani sm s occ a sionally hav e e f fec ts o n risk a nd s o could poten tial ly be incl ude d. The approac h we have t a ken t o a ddre s s t hes e i ssue s i s to c arry out wei ght ed burd en a naly se s, in whi ch v ariants judged a priori t o be mo s t lik ely t o have importa nt e ff ect s ar e a cco rded higher we ights. Sinc e sel ec t i on pr e ssure s me an that c ommon v arian t s ar e unlik ely t o hav e large e f f ec t s , va r i ant s a re al so wei ght ed ac cording to ra r i t y and t h e de tailed s c hem e fo r doi ng t his i s de scrib ed in the Me thod s sec tion. How ever a wea kne s s of t hi s a pproa ch to d a te h as b een t ha t t he r e h a s been little e mpirica l evi denc e t o in form th e ex a ct w eighting s c h eme w hich would be op timal . An adv antage o f th e large U K Biob ank data s et i s tha t it allo w s s ome explo ra t io n of t h e rela tiv e av er a g e All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint ef fec t s iz e s of di ff er ent ca t e gori e s of vari a nt and t his wa s c ar r i ed ou t u sing mu lt i variat e a naly s e s o f va r i ant ca teg ori es in a ddi t io n to s ta nda rd weig hted burd en a naly se s o f gen e s a n d gen e se ts. Th e se inv es t ig ation s wer e a ppli ed to t he pre viously u sed hyperli pid aemia ph eno type, de fined a s subje ct s wi t h a dia gno s i s o f hype rlipid aemi a and/ or t a king chole st erol -lowe r in g medic ati o n.

Methods

The U K Biobank data s et wa s dow nload e d along with the v aria nt c all fil e s fo r 20 0, 632 subjec t s w ho had undergon e exome -s equ enc ing and ge notyping by the UK Bio b ank E xome S equenc ing Con sor t i um u s i ng the GRCh38 a ss embly with cove r a ge 20X at 95.6% of si te s on a ve r a ge (Szu s ta kow ski e t al., 2 020 ). UK Bioba nk h ad obtain ed e t hi cs a pprova l from th e No r th We s t Multi - ce ntre Re s e a rch Ethic s Commit te e wh ich c overs th e UK (app rova l number : 1 1/N W/ 038 2) a nd ha d obtain e d info rmed c on s e n t from all pa rti c ipant s. The UK Bi ob ank a pproved a n ap pli cation f o r u se o f the da t a ( ID 51119) a nd ethic s a p pr o val f or the an aly se s wa s ob t a in ed f rom the U CL Re s e a rch Ethic s Committ ee ( 1 1527/001) . All varian t s we re annot at ed usi ng t he s t anda r d s o ftwa r e pack ages VEP, Poly Ph en a nd SIF T ( Adz h ub ei et a l ., 2013 ; Kumar et a l ., 2009 ; McLa r e n et al ., 2016 ). To obtai n popul ati on princi pal c ompon ent s r efl ecti ng anc estry, v er s io n 2.0 o f plin k (ht tp s: / / www .cog - gen omics. org/pl ink/2.0 /) wa s run w it h th e option s -- ma f 0.1 --pca 20 appr ox (Cha ng e t al., 2 015; Galin sky et al ., 2016). The hyperlipi da emia pheno type wa s de t ermine d in th e same way as p reviou sly f r om fou r sourc e s in the da t a s e t: sel f- rep orte d high chole s t er ol; repo rting t ak ing c hole ste rol low eri ng med ication ; repor ting tak ing a named stati n; ha ving a n ICD10 d i agno si s fo r hyperlipi da emia in hospi tal r ec ord s or as a c au s e of d e ath (Cu rt i s, 20 20 ). Subj ec ts in any o f th e se ca te gorie s wer e de eme d to be ca se s with hyp er l ipida em ia wh ile a ll oth er subject s were t a k en to be con trol s . The method o f an aly si s was the same as u s e d pr e viou sly o n th e small er sample . The S CORE A S SO C program w as u se d to ca rry o ut a we ight e d burden ana ly s i s to te s t whethe r, in eac h gene, s eq u ence va r i ant s w hich were ra rer and/ or pre dic te d to hav e mor e s ever e func tional e f f e ct s occ ur r ed mor e co mmonly i n ca s e s than c on trol s. A t ten ti on w as re strict ed t o r a re variant s wi t h m inor alle l e freq uenc y (MA F) < = 0.01 i n bot h ca se s a nd control s . A s prev iou sly de sc r ib e d, va r i a nt s w ere we igh ted by overall MA F s o tha t varian ts wi t h M AF =0.01 were gi ven a weig ht o f 1 w hile v er y rare va r i an ts w it h MAF c los e to ze r o wer e gi ven a wei ght o f 10 (Cur tis , 2021). V a r i ant s w er e al so w ei ghted acc or d ing t o their func t i onal a nno ta tion u sing the GE NEVAR ASS OC prog ram, whic h wa s u sed t o g enera te in pu t file s f or we ight ed burd en analy s i s by SCO REASSO C (C urti s, 2016, 2012). A ma ximu m w eigh t o f 40 wa s a ssigned to varian ts p redic te d to ca u s e compl e t e LOF o f th e gene , n amely sto p-ga ined , frame shi f t and e s sen tial splic e sit e va r i an ts. O ther type s of va r i ant wer e a ssigne d i nterme diat e we ights, fo r examp le, a weig ht o f 5 wa s a ssigned for a synonymo u s v arian t, 10 fo r a non- synony mou s v arian t an d 15 fo r infr ame i ns e rtio ns a nd dele tion s . Additi ona lly, 10 wa s a dd ed to t he we ight if th e Poly P he n ann ota tion w a s p ossibl y or pr obably dama ging a nd al so i f the SIFT ann ota tion wa s d el ete riou s, me ani ng tha t a n on - s yn onymous va r i an t anno ta ted a s bo t h dam agi ng and dele te r i ou s w ould b e a ss ig n ed an over all w eigh t o f 30. In o r de r to a llow explo rati on of th e e ff ec ts o f diff er ent typ e s o f v arian t on d i s e a se r i sk the varian t s we r e al s o grouped i n t o br o a der ca te gorie s to be u sed in multiv aria te a na lys e s a s de s c ri bed bel ow. T he full se t of w eight s a nd c a tegori es i s display e d in T able 1. As d esc ribed pr ev iou s l y, the w eigh t d ue t o MAF an d th e we ight due to f unc tion al annota tion w er e multipl ied t ogeth er t o prov ide an ove rall weig ht for e ach variant . V a r ia n ts w ere e xc luded if the re we r e more than 10 % of ge no type s mi ss i ng in the contr ols or if the he te rozygote c ount wa s small e r than bo t h homoz y gote c ou nt s i n th e con trol s. If a s ubje c t wa s no t g eno typed fo r a varia nt t h e n th ey we r e a ssign ed th e subjec t-w ise ave r a g e s core fo r tha t va r i an t. Fo r ea ch subjec t a g e ne -wis e All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint we ighted burd en s c ore wa s de riv ed a s th e sum o f the v ari ant -wi se w eight s, eac h multipl ied by the numbe r o f allele s o f th e varian t w hich t h e giv en subjec t po ss e s s ed. F or varia n ts o n the X ch r omo some, hemizygou s ma le s wer e tr eat ed a s hom oz yg ote s. For e ach g ene , a ridge r eg r e s sion a na lysi s wa s ca r ri ed out with l amd a=1 t o t e st wh ethe r t h e gene - wi s e varian t burd en s c or e wa s a s socia te d with the hy perlip ida emia p h eno type. T o do thi s, SCOREA SS OC f i rs t cal cula te s th e l ikeli hoo d for th e phen o t y pe s a s pre dict ed b y the fir st 20 popul a t i on princ ipal c omponen t s and th e n cal cula te s th e l ikeli hood u sing a mod el w hich additiona lly inc orpor a t e s t he gene -w i s e burd en s c or e s . It t h en ca r ri e s out a like li hood r atio te s t as s umi ng t hat twic e the na t u ral log o f t h e like lihoo d rat io f o llow s a c hi- squ are d dis tr i bu t i on with o ne deg r e e of fre edom to p roduc e a p val ue . The s ta ti s tical s i gni fic ance i s summa r i se d a s a s i gn ed lo g p v alue (S LP) whi ch is the l og ba s e 10 of th e p v alue given a posi tive s ig n i f th e sco re i s hi gher i n c ase s an d neg ative i f it i s higher in cont rol s. I n prev i ous an aly se s it app e ar ed tha t inc orpo r a ti ng popul ation princ ipal c omponen t s in thi s wa y sa ti sfa c torily c ontr olled for t e st s t ati s t i c infl a tion whe n appli ed to the a nce s tr a lly he t e r og ene ou s UK Bioba nk data se t (C urti s, 2 021). Howe ve r preli mina r y ana lys es o f this n ew, large r dat as et r eve ale d tha t th ere wa s a sligh t ten dency fo r more r are v aria nts in X ch r omo some gene s t o be i den ti fied i n fe males rath er th an ma le s. H ence, s ex wa s a ls o inclu d ed a s a co variate a l ong w ith the p rinc ipal c ompo nent s and t hi s produc e d a well-b e haved t e s t sta ti stic , a s d e t a ile d i n t he R es ul t s s e ct i o n . Gene s et ana lyse s we re c arrie d ou t a s be for e usi ng t he 1454 "all G O ge n e s et s , g e ne symbol s " pathw ays as li s ted in th e fil e c 5.all. v 5.0. s ymbol s.gmt do wnload e d f rom the Molec ular Si gnatu re s Dat aba se a t h ttp: // w ww. br o a din st i tut e. org/g sea /m s i gdb/coll e ction s .jsp (Subram ani an et al ., 2005 ). For e ach se t of g en e s, th e natu ral l og s of the gen e -wi se p v alue s we re summe d ac c ording to Fi sher ’ s

Method

t o p r odu ce a chi- squa red st ati sti c with d egree s o f fr eed om e qual t o twi ce the number o f gen e s i n the se t. The p val u e as soc iat ed w ith thi s chi- squa red st ati stic wa s e xpr essed a s a minu s log 10 p (MLP) a s a te st o f a ssoc ia tion o f t he s et w it h th e hype r li pi daemi a ph eno ty pe. For s elec t ed ge ne s, additi onal anal y s e s w ere ca rried ou t to cla r ify th e c ont rib utio n of dif fe ren t ca tegori e s of va r i a nt. To do t hi s, e ach c at egory as li s t e d in Ta ble 1 wa s a s s ig ned a weig ht consi sting of a di ff eren t power o f 1 0 and th en G E NEVARAS S OC and SC OREASS O C we re u sed to obtai n s c o re s for e ac h su bject a s th e s um o f t h e s e wei gh t s . Thi s a ll owe d th e ove r a ll numb er o f v ariant s o f eac h ca tegory po s se sse d by a subjec t t o be c o ded a s a decima l numbe r s o tha t, f or ex a mple, a sco re o f 1000 302 w ould indi cate tha t th e s u bje ct pos ses s ed one o f on e c atego ry of v arian t , thre e of a noth er ca tegory a nd two o f a t hird cat eg or y . C o de was writ ten in R t o r e ad in t he se sc o res an d par s e t h em to obt ain t h e s u bjec t-wi se c oun t s for e ac h c atego r y o f varia nt (R Co re Tea m, 2014 ). The se we re t h e n ente red i n t o a log i s tic reg re ssion an aly s i s of c ase - c ont rol s t a t u s al ong wi t h pr i ncip a l com ponent s an d sex i n ord er to estima te th e rela tiv e con tribution s o f di ff eren t v arian t c atego rie s t o the phe n ot y pe . The odds r a t i os a ss ocia te d with the ca te go r y were e s tima ted al ong with thei r s ta ndard er ror s a nd the Wal d stati s tic w as u sed t o ob tain a p val ue, ex cept for c at egori es in w hic h vari an ts oc curre d few er t ha n 50 time s in whic h ca s e Fi she r’ s e xac t te s t wa s a ppli ed t o the raw v aria nt co unts w ithou t inc luding cov ariate s. The a ssocia te d p va lue w as conv erted t o an S LP, a gain w ith t he sign bei ng posi tive if t he mean coun t w a s higher in case s than c ont r ol s.

Results

R e su lts of g e ne-w ise w eigh ted b urd en tests The r e wer e 44, 054 c ase s w ith a di agno si s o f hyperli pida emia and /or taking chol e s t erol -low e ring medi cation an d 156, 578 c ontr ol s . Th ere were 2 2,642 g ene s for whic h the re we re qualify ing v arian ts and pr e limina ry analy se s sh owed tha t th ere wa s a bia s towa rd s produci ng strongl y neg ative SLPs, All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint whi ch w as confin e d to gen e s on t he X c h r omo some. The a nalys e s were r epe ate d using sex as a pheno t y p e and t h i s c on firmed tha t the fr eque ncy of ra r e , damag ing v ariant s wa s hig her in fem al e s for gen e s on t h e X c hr om o s om e. Thi s woul d occ ur if th e ge notyp e c allin g alg or i t h m were slig htl y more li kely to cal l a f em ale a s h et erozyg ous than a male a s h emizygou s. S ince th e fr equenc y of c a se s is low er in femal e s , t h e overal l e f fect i s t o obs erve an ex ce s s o f r a re , damag ing v ariant s in c ontrol s rath er tha n c as e s fo r g ene s on t he X chro mosome . The refo re the analy s e s w ere re peat ed fo r hyp er l ipida em ia usi ng sex a s a c ovaria te a s well a s t h e principa l co mpo nen ts. W h e n thi s was d one only two ge ne s produce d strongly po s i t i v e or neg a tive SLPs, LD L R (SL P = 5 0.08) a n d PCSK9 ( SL P = - 10.4 2). The qu antil e-q ua nt i l e ( QQ ) plo t f or the S LPs ob tain ed f or each o f th e rem aini ng ge ne s is shown i n Fig ur e 1 . Thi s show s tha t th e te st a ppea rs to be well -be haved and c on fo r ms f airly wel l with the ex pec te d di stributi on. O mi tting th e gene s w ith th e 100 highes t and 100 low e st S LPs, wh ich mi ght be c apturi ng a real bi ol ogic al e ffe c t, th e gradi ent for p os i tiv e SL Ps i s 1.08 w ith int ercep t a t - 0 .019 a nd the g r a di ent for n ega t iv e SL Ps i s 1.04 w it h interce pt a t -0.013 , i ndica t i ng only mo des t infl atio n o f t he te st st ati s t i c. T h e ro l e o f v e ry r a r e v a r i a n t s i n b o t h LDL R and PCS K9 in the pat hogene s i s o f f ami l ial hy perlipida emia is alr ea dy we ll e st abli shed . How ever th e re sult s f rom th e curre nt anal y s i s i mplica te a la r g er numb er of v arian t s i n the se gene s ha ving a range of e ff ect s on ri sk o f hy perlipid a emia i n th e popula t i on more gen erally. The s e ar e pre se nte d in det ail b elow i n the de sc ription o f th e re s ul ts of the a naly si s o f ef fec ts o f di ff eren t v arian t c at egorie s . Given t h a t th er e wer e 2 2,642 i nf or ma t i v e ge ne s, th e critic al thre sh old f or th e ab s olute v alue o f th e SL P to dec la re a re sult a s formal ly sta ti s ti c ally s ig nifica n t i s - l og10(0 .05 / 22 642 ) = 5 . 66 and t hi s w a s ac hieve d by thre e ot h e r gen es , AN G PT L3 (S LP = -5 .67) , L O C1 0272372 9 (SLP = -5.7 7) and IF ITM5 ( S LP = -5. 86 ). Lo s s of functi on varian ts in ANGP TL3 have pr e viou sly be en shown to c au se c ombin ed hyp olipidae mia a nd i t is the targe t of ev inac umab, a human monocl on al an t ib ody design e d to t r e a t hyp er c hole s terol aemi a (Doggr ell, 2020; Wang and Mu sunur u, 201 9). H owev er IF I TM5 a nd LOC102 723729 do not se em to be biol ogi call y pla us ib l e ca ndida te s. IF IT M 5 i s invol ved i n bone mine r a li s a t i on and v aria nt s in it a re a kn ow n ca us e o f o s t eo gen e si s imper fect a (Why t e e t al., 2020 ). LOC102 723729 is a poo rly charac t er i s ed l ncRNA whi ch ma y ac t as a tumour supp r esso r in non - small ce ll lung ca ncer (Ya ng e t al ., 2019 ). A to ta l of 55 g ene s had SL P with ab solu te v alu e of 3 or mor e (equiv ale nt t o un corre cted p = 0.001 ) , w herea s o ne would only e xpec t a round 22 64 2/ 1 000 = 2 3 by ch ance s o a number o f th e se ma y in f a ct be exerting some ef f e c t on ri sk. T he s e a r e lis ted in Table 2 and s om e appe ar to b e of p articul ar int e re s t a nd a re di scu s sed bri e fly as bel ow . The re sul ts for al l gen e s a r e p re s e n t e d in S uppl emen tary Table 1. G6PC (S L P = 5.55 ) is o f int er es t beca u se mutation s ac ting r e ce s s i vely cau se g lyc o gen st or a ge di sea s e type I ( G S D1 , von Gie r k e dis ea se , inc iden c e ~1/100, 000 ), wh ich i nclude s hy perlipi da emia a s pa rt of the phe n otype (Kish nani e t al., 2014 ). Ra r e homozygou s va r ia nt s in A BCG 5 ( S L P = 3 . 3 1 ) ca n pr o du ce sit o s te rola e mia and ar e a k nown caus e o f hom oz y gou s familia l hyp e r c hole s te r o la emia (C uchel e t al., 2014 ). Muta tio ns in ABCD1 (SL P = 3.2 6 ) cau se X -linked a dren oleuk ody stroph y, wh ich resu lt s in ele vat ed l evel s of v e r y long chai n fa tty ac ids in pla s ma and ti s s ue s (Eng el en e t al., 20 12). V arian t s in GCK (SL P = 3 .04 ) ar e k nown to c au se mat urity-o ns et di abe t e s o f the young ( M O DY) wi t h mild hyp er g lyc aemia and lo wer tr i glyc eride l e vels th an oth er f orm s of type 2 diabe t e s (Ma e t al., 20 19 ). Li ke ANGPTL3 , AN GPTL 4 (SL P = -3.6 6 ) modula t e s the a ctivi ty of lipopr ot ein lipa se ( LP L) and ina ctiva t in g va r ia n ts in it have pr eviou sly been shown t o be as soc iat ed with hyp ol ipidae mia ( Dro n and H e gele , 2016). Th e p roduct o f NP C 1 L1 (SL P = -3 . 70) i s e sse ntia l for in te s ti na l s t e r ol a bso r p tion and is the molec ula r ta rget o f ez etimib e , a poten t c hole s tero l ab sorp t i on inhibi t or that l owe rs bl ood ch oles t e rol (B ette rs a nd Y u, 2010 ) . H o m oz yg ous knoc kout o f PPP 1R 3 G ( S LP = -4. 25) mi t ig ate s high - fat di e t induc ed obe si ty in mi ce (Z hang e t al., 2 017) . V arian t s in AP OC3 ( SL P = - 4 . 89 ) h a ve p r e v io us l y shown t o be pro t e ctive agai n s t hy perli pi dae mia ri sk ( Dron an d Heg ele , 2 016). All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint While t w o AT P bi nding c as set te t ran spo rter g en e s, ABC G5 and ABC D1, p r o duc ed S LPs ab ov e 3, a third, A BCA1 (S L P = - 2 .91), wa s o nly ma rg inally les s sign ifica n t . The p roduc t o f AB CA1 is re s p on sible for t r a n s po rting chole ster ol out o f c ell s a nd homoz y gous or com po und he ter ozygous va r i a nt s i n it ca use Tang ier di sea s e, a fa milial H DL de fi ci ency s y ndrome, w hile h e te r ozygo u s v ar iant s ar e as s o cia ted with reduc e d HDL lev el s (M ar anghi et al. , 2 019; Pun toni e t al. , 2012). I t ha s a n e sta bli shed role i n th e regula t i on o f HD L and t he re ar e repo rt s th at c ommon v arian t s in it ar e a ssoci a ted wi t h pla s ma lipid l evel s (Kolda mova e t al. , 2 014; L u et al. , 2018 ). R e su lts of g e ne s e t an a ly s es An ini t i al ru n of t he g ene s et ana lyse s t e n ded to hig hlight se t s cont aini ng hundr e ds o f ge ne s whic h inc luded o ne o r more o f th e g ene s with a bs ol u t e SL Ps ov e r 3 as li s ted in Table 2 so the ana ly se s wer e repea ted wit h the s e ge ne s and A BCA1 o mi t ted t o see i f any additi ona l gene s o f i nt e re st co uld be iden t i fi ed . Given tha t 1,454 se t s were te s ted a crit ical MLP to a chie ve to d ec lar e r esul t s sig nifica n t aft er co rrec t i on fo r multiple te sti ng w ould be log10 (1454*20 ) = 4.46 and t his wa s not a chiev ed by any s e t . In s p e ction o f th e r e sul t s fo r t h e highe st sc oring set s did no t rev e al a ny a d ditional g en e s whi ch mi ght obvio us ly be in volve d in hy perlipida emia ri sk. T he r es ults f or al l s e ts are p r ov id ed in S u pp l e m e nt a r y T a b l e 2 . Res ult s of variant c at egory analyse s For the t w o gene s showi ng the mo s t def i nite evide nc e of a ssoci ation , LD LR and PC S K 9 , a logi s tic regre s s io n an aly s i s o f di f f e r en t c at egorie s o f v ariant w a s c ar ried ou t to e l ucida te th eir rel ative co nt ribu tion s. T he r e s u lts for LDL R a re sh own i n Tabl e 3A. I t c an be s ee n tha t dis ruptiv e va r i an ts, co mpr i s ing s to p v ariant s an d fr ame shi ft va r i ant s , a r e s i gnific an tly a ss ocia te d with c asen e ss (S LP = 16.9 5) with a larg e e f f e c t on ri sk ( OR = 4 0 .02 (11.83 - 135 .33 )). Ther e we r e 34 o f t hes e v arian t s , of whi ch onl y 3 w ere see n in control s . Es se ntia l s p lice s it e v ariant s al s o ex er t ed a lar ge ef fect o n r i sk (OR = 10 .4 (1.9 - 56. 7) ) , SL P = 5.5 5, with 1 1 out of 13 being se en in c a s e s. S t op va riant s, fram es hif t va r i ant s a nd e s sentia l splic e varian ts a re e xpec t e d t o ca u s e LOF bu t the r e sult s sh ow that o ther va r i ant s w hich do not seve rely di sr u p t th e gene but whic h prod uce cha nge s in am i no ac id s equenc e also h a ve mode ra t e ef fec ts on ri sk. The re were 6, 747 non synon ymou s varia n ts a nd t hi s c atego ry was as s o cia ted with OR = 1.1 5 (1.05 - 1 .25) . Howev er of the se 1,175 wer e a nno ta ted b y SIFT as “ de l e t er i o us ” an d t h is c a te g or y h as O R = 1 . 7 4 ( 1 . 4 1 - 2 .1 4 ) w h i le t h e r is k ass oc i at ed w i t h a n anno t a t io n by Poly P hen o f “probab ly da m aging ” w as s mal ler , 1.30 (1 .03 - 1.6 5 ), a nd ther e w as no signific ant ri sk a s s oc ia ted with an annot a tion of “ po s s ibl y dama ging” . In frame in sertio n/d ele tio n va r i ant s w er e ob served on 10 oc ca sion s and d eta iled in s pe c tion o f th e r e sul ts rev eal ed t hat the se co ns i st ed o f de le tion s a t 4 dif fer ent po si t ions, one o f whic h occ urr e d in 7 dif f er e n t s ubj ec t s . All 10 o f the subjec ts w ith on e of the se d e le tion s was a c a se ( SL P=6 .58 ). By c ontr a s t, ge ne t ic v ariant s whic h did not a ff ect pro tein se que nce i n gen er al did not ha ve signi f i c ant e f fec t s on r i s k . The ex ce pt i on wa s that the “ Splic e Reg ion” cat eg or y se eme d to e xert a prot ectiv e e f fec t, with OR = 0. 86 (0.81 - 0.92), SL P = -5. 21 . Thi s w a s dr i ven by r s 72 6588 67, whic h had frequ e ncy 0.012 i n control s and 0.0097 in ca s e s a n d whic h has b ee n pr eviou sly r e p or ted t o be a s s o ci ate d with low er chol e sterol a nd low er r i sk of c or o n ary art ery di sea s e (G re tarsdo tti r et al. , 2015). W hen t h e anal y s i s w as r e pe ated w ith thi s va r i ant re moved, t h er e wa s no g en eral t end ency fo r s pli c e reg ion va r i a nts t o be a ssoc ia ted w ith r i s k (OR = 1 .13 (0.99 - 1. 29) , SL P = 1 .20 ). Tab le 3B sh ows t he re su lt s o f the va r i a nt categ or y ana ly si s f or PC SK9 . I t c an be s ee n that t he s e a re broadly simil ar to tho se ob tain e d fo r LDL R, albei t in the opp o site di rec t i on bec au se impai r ed func t i o n of PCSK9 reduce s ri sk of hy pe rli pida emia. Di srup tive, e ssen tial s pli ce si te and missen s e va r i ant s a nno ta ted a s “d ele te riou s” by SIF T are al l signi fic an tly mo r e common in co nt rol s an d have All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint an overall O R o f a r o und 0.5 . I t i s i nte res ti ng to n ot e that the se categ o r i e s o f varia nt oc cur more freq uen tly in PCS K9 than i n LDLR . In LDL R there are onl y 34 di s rup tive varia n t s wh er e a s in PCS K9 ther e are 291 and in LD LR ther e ar e onl y 13 essen tial s pli ce si te v arian t s while in PC SK 9 there ar e 193. The se re sul ts allow us t o gain some in s i g ht into t he ov er all im pact o f v ariant s in t hes e gen e s o n the risk o f hy perlipid ae mia in t h e gene ral po pulatio n . For va r i a nt s i n LDLR whic h are nonsyno nymou s and annot at ed a s “ delet eriou s” by SIF T, t he o verall e stima ted O R i s 1. 15 *1.74 = 2. I t sh ould be empha s i sed th a t t his e stima te i s fo r the av erage e f fect o f such variant s a nd tha t t here i s l ikely to be co ns id e r a bl e va r i ati on, w ith s ome o f th e se vari a nts ex er t i ng marke d e f fect s on ri s k wh ile other s may hav e trivial e ff ect s or may even b e prot e ctiv e. Ther e a r e 889 o f t h e se v arian t s an d, sinc e they are rare , few pe ople have mor e th an one o f them s o tha t we can say that around 8 50 out of the 2 00,000 subjec ts , or slightly le s s tha n 0 .5%, have a del ete r i o us v ari ant in LDLR w hich , on a v erage, abou t doubl e s th e odd s of hyp erlip ida emia . Thi s co mpar e s with the 47 LOF v aria nt s whi c h confe r h igh ri sk but w hich o ccur i n only 0.02% of s u bj ect s. Del ete riou s varia n ts in PC S K 9 on ave r a g e hav e OR of abou t 0 .5 a nd occ ur i n 0.8% of subject s whil e LOF variant s have a simila r OR a nd occ ur in 0.2 % of subjec ts . Bro adly s p e akin g, it se ems that about 1. 5% o f pe opl e w ill ha ve a r a re co ding varia nt in one of th e se two gen e s whi ch ei t he r d ouble s or halv e s t he o dd s of d eve loping hyperli pi daemia . R e su lts fo r s e l e c t e d g en es It i s rel ev ant al so t o rep ort ce rtai n g ene s whic h produc ed neg a tive re sul ts . Wi t h t h e exc eption o f LDL R, none o f t h e ge ne s hi ghligh t e d by the pr e viou s ana ly si s o f 50,000 UK Bi oban k e xomes s howed any evi dence fo r a ssoci ati on in thi s en l ar ged sampl e once s ex wa s i nclud e d a s a c ovariat e. The se g e nes co n s is t of HUWE1, CXorf56 , RBP2 , ST AT5B , NPFF R1 , AC OT9 , GK, ADIPOQ , SU R F 1 , AD RB3 , GYG2 , PHK A 1 and PHKA2 (Cu rt i s, 20 20 ). HUWE 1 and a number o f o ther s a re loc a t ed on the X ch r omo some and w it h hind sigh t it ap p ea rs t ha t the y may have produce d strongly nega t i ve S LPs a s a co ns e qu ence o f th e red uce d f r e qu ency of va r i ant s c all e d on th e X chromo some in m ales, whi le oth er re sult s ma y hav e simply been due to cha nce. Other ge n e s fo r w hich notably nega tiv e re s ul ts a re obtain e d are AP OB (SL P = 0 .0 0) a k nown c aus e of famil ial hy perch ole s t erol ae mia, an d HMG C R ( SL P = - 0 . 0 7) , w h i ch c o d e s f or t h e r a te - l im it i n g e nz y m e i n ch o l es t er o l s yn t he s is wh i c h is t h e t ar ge t of s ta t i ns (La Ro s a e t al. , 1999). Al so n eg ative wa s STAP1 (SLP = -1.02 ), for w hich t h e r e wer e in itial c laim s of an as s o cia tion with fami lial h ype rchol e stero lae mia a lt h ough more r ec en t work has throw n doubt on this (Kanuri et a l ., 202 0). Th es e th ree g en es wer e a l s o s ubje ct ed to the v aria nt c at egory a nalysi s an d no c ategory o f varian t w ithin th em show ed signi fic ant as soc iatio n with hyp er l ipid ae mia. Dis c us sion The se a naly se s provide a broad ove rvie w of c ontr i b ut i on s o f r a re coding gene t ic va r i ant s to th e r i sk of hy perlipi dae mia . T her e are a numb er of i ssue s w ort hy o f fu rt h er commen t. The obse r v ati on tha t in t hi s da t a s e t rar e X chromosome va r i ant s a re cal led mor e f reque n t l y in femal es tha n in ma le s i s impor t a n t t o re c ogni s e . U nle ss t his ef fec t is a ll owed for, f or ex ample by inc orpor a t i ng s ex a s a cov ari at e, a rte fac tu al re sult s ma y be prod uced for a ny phe n ot y pe w ho s e preva lenc e vari e s w ith s e x. Wit h hind sigh t , t hi s oc cur red in t h e ea rlier a n alysi s o f the 50, 000 ex ome s and led to the i de nti fica t i o n a s some gen e s on t he X ch r omo some a s b eing p ote nti ally r e levan t. Going fo rw ard, r e s e a r c her s ne ed to b e a wa r e of thi s p henom enon and d ea l wi t h i t approp riat ely. Working with biob ank da t a s e t s ca n po se particul a r chal lenge s c ompa red t o tr aditi ona l ca se -c ontrol st u d ie s . In a c a se -cont rol study one can c ontrol r ec r ui tme nt a nd a s s e ss su bject s a ga inst pre - speci fied criteri a. W i t h t h e U K Biobank one h a s a se lf -sel e cted sampl e of volun t e e rs al ong wi t h informa tion abou t a br oad ra ng e of ph eno ty pe s but some mea sure s a r e only avai labl e f o r a su bse t o f the sampl e. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint The phenotyp e studie d her e is in tend ed t o broad ly capture c linic ally s i gnific a nt h yperlip idaemi a , using a s i t d o e s a com binati on of the dia gno s i s and the mo s t c ommonly use d t r e a tment s. H ow ever this p heno type cle a rly diff er s f r o m what on e might u se i n a more sy st ematic ally a sse s s e d s a mpl e. N o att empt wa s m ade t o inc orpor at e ac t ua l mea s u re s o f blood lipid s, in p a r t b e cau se the se mig ht be dist or t ed by tr eatme nt e f fec ts . S ome sub jec t s wi ll hav e be e n pre s c ri bed sta tin s p u r e ly on the ba si s of rai s ed li pid s fou nd during r ou tine cli ni cal as s e s sme nt wher ea s oth er subjec ts w it h some what low er leve l s m ight be r ec eivi ng them be c aus e they had c a rdiova s c ular di se a se. L ik ew ise , s ome subjec ts cl a ssi fied a s con trol s might in fa ct hav e hyp er l ipida e mia w hich h a s not b een di agno se d. Thus, t he p hen otype i s un d ersto od to b e a quite n oi s y and a di sta nt con se quenc e of th e i mmediat e biol ogica l ef fec ts of a ny func t i on al ge ne ti c v ariant s. Ano the r issu e i s tha t th e parti c ipan ts rep re s e n t a relativ e ly h ealthy group o f subject s . Pe o pl e with s ev ere hyp erli pida emia whi ch ha d re s ul ted in ea rly dea t h would no t be inc lud ed, meaning th at th e ef f ect size s ob s e r v ed in thi s samp le may tend to b e undere s timat e s. The curren t anal y s i s hi ghlig hts a number of ge ne s for w hich very rar e v ariant s wit h la r g e e ff ec t size hav e previ ou s ly been s how n t o impac t li pid leve ls a nd demon st ra te s tha t l arg e numbe rs of addi t i on al v ariant s with mo re mo de r a te ef fect a lso m ake a broa der c on t ribu tio n t o ris k in th e gen eral pop ul ati on. Thi s is mo s t cle arly t he c as e fo r LDL R and PC SK9 but th er e a re a few o f oth er gen e s w hic h probably al so show thi s e ff e ct, e s pe cial ly AN G PTL3 a nd ANGPT L4 . Conv e r s el y , o t he r gen e s w hic h are impl icat ed a s monogeni c c ause s o f sev er e famili al hy p erlipida e m ias, s u ch a s APO B and STAP1 , are not id enti fie d by thi s ap p r oa ch a s ma king br oa de r cont ributio ns t o hype r l ipida em ia risk. Th e analy s e s hi ghligh t th re e gene s whi ch are alre ady the t arge t s of lipid -lo w ering the rap ie s, PCS K9 , ANGPT L3 a nd NPC1L1 , but comp l etely fa iled to de t e ct an e ffe c t fo r HMG C R , w hich encod e s the t arget o f st atin s. Th e appro ach use d i s int ended to de t e ct the add i t i ve e ff ec t s of varian t s whic h are i ndividu ally very rar e bu t w hic h cumul atively hav e an e ffe ct on the fun ction o f a gene . H e nce i t is not ex pect ed t o b e succ e ss ful i f the ef fec t of some v ari ant s impairing gen e f u nc tio n ma y be co unterba lanc e d by ot h ers w hich pr od u ce a gain of function . I t is n ece s sary to gro up v ariant s bec au se whe n a v aria nt i s only ob s e r v ed in a handful o f s ub ject s it i s no t po s s ibl e to draw fi rm co nclusion s ab out i ts ef fec t . Ther e mig ht be scop e t o gain powe r by devising more s o p histic a ted approa ch e s to va r ia n t cla s sifica t i on, for e xam ple rela ted t o t h e mor e sp eci f i c pr ed ic tion s abou t ef fec t o n the p rot ein produc t. Estima ting t he ef fec t on r i sk o f di f fere nt categ or i e s o f vari ant w ithin LD LR an d PC SK9 broa dly co nfir m s w hat we might hav e ex pec ted . LOF v ariant s , compri sing sto p, f rame shi ft and splic e si te va r i ant s , h ave l a rge e ff ec t s . Va ria nts c a te goris ed a s “d elet erio us” by SIF T hav e mo dera te e f f e ct s on risk, w he r e a s t h e ca teg ori s a tio n a s “prob abl y dama ging” by PolyPh en is a s s oc i ate d w ith a somew ha t smalle r e ff ec t. The annot atio n of “ po s s ib ly dama ging” does not se em to ha ve mu c h utili t y in thi s co nt e xt. The analy s e s show tha t e ven n o ns y nonymou s va r i an t s i n LDLR whic h do not ha ve a ny of the se an not atio n s a r e s till, on av erage , a ssoci a ted with a s l ightly incr ea sed ri sk, w ith OR = 1.15 . Vari ant s whi ch do not cau s e c hange s i n a mino a cid s eq u ence do n ot in g ene ral se em to exer t an apprec iabl e e ff ect on r i sk. Th es e find ing s w ill be of us e in c on st ruc t i ng we ighting sc hemes for future ana lys e s of th is n atu r e . The v ari ant s a re mostly al l indiv iduall y ex treme ly rar e but i n tot a l t her e are far mor e non syno nymou s varia n t s than L OF va r i ant s. The general pic ture w hich eme rge s is tha t the re i s a rel a t i vely s m a ll numbe r o f ge nes in w hich va r i ant s w hich are indiv idual ly ex treme ly rare ma ke a n a pprec ia ble cont ributi on t o the o ver all ri sk of dev eloping hyperli pid aemia . Few va r ia nt s c au se LOF bu t t h o se w hich do hav e a l a rge ef fec t, wher ea s far la rge r numbe r s o f no ns y no nymou s v a riant s te nd to exe r t mor e m oder at e ef f ects . Neve rth ele s s, the c umulativ e fr equenc y of the s e va r i an ts r ema in s low. I f we c on fine a t te ntion to t h e r esul ts abou t whi ch w e ca n feel mo s t c on fiden t, it s ee ms tha t f ew er t h a n 2% of pe ople car ry a variant w hich might hal ve or doubl e ri s k . It wi ll be p o ssibl e to refin e e st i ma te s suc h a s thi s a s more da ta be come s av ailab le, for ex ample from the rema inin g 300,000 U K Biobank s ub ject s for whom e xome sequ e nce All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint data i s y et t o be prov id ed. Wi th a large r data se t it w ill bec ome p o ssibl e to d raw more de fi nitive co nclusion s ab out i ndividua l ge n e s a nd t o mak e more a ccura te e stima te s of e ffe c t s i ze s. The ava ilabili t y o f s equenc e data from a l arge number o f subjec ts ha s al l owed in s i g hts in t o t h e co nt ribu tion whic h rar e co ding g enetic v ariant s ca n make to hyperli pi daemia , an i mpor ta n t pheno t y p e whi ch is a l s o a ssoci a ted wi t h a varie t y of socio ec onomic and e nvironm enta l ri s k facto r s . More de tai l ed ana ly se s may foc u s on sp e cific gen e s a nd / o r v aria nt s, may inve stiga te sign al s of sel ec t i on p re ssur es, may look a t int er a c ti ons b etwe e n dif fe rent ge n etic a n d envi ronme ntal va r i able s and may e xplore th e devel opme nt o f ind iv idual is e d r i s k ass e ss m e nts . Hype r li pida emi a provide s a use ful pa rad igm of a common compl ex tr ait and s im ilar app roac h e s ca n be ap pli e d to oth er pheno t y p es. Conflic t s of interest The autho r dec lar e s he ha s no c on flic t o f inter est. Data availability The r a w da t a i s avai labl e on appli ca tion to U K Bi obank. De t a iled r es ult s with va r i a nt coun ts c anno t be m ade av aila bl e bec au s e they might b e used fo r s ubje ct iden ti fic ation . Re l evan t d er i ved v aria ble s inc luding principa l co mponen t s and va r i a nt anno tati on s w ill be depo sit ed in UK Bi ob ank. S crip ts a nd so ft w ar e u sed t o c arr y out the an a lys e s a re ava ilabl e at http s : / /github .co m/ da ve nomidd len amec urt i s . Ackn owledgme nt s This re s earch ha s b een c onduc ted u s in g t he UK Bioba nk Re sou rce. The a ut h o r w is hes t o ac know ledge t h e staf f suppo rting th e H ig h Per forma nce Compu ting Clu ster , Co mp uter Sci e nce Depa rtmen t, Unive r s ity Col l ege London . This work w as ca rried ou t in par t u sing re sou rce s provide d by BBSRC equ ipm ent gran t BB /R0135 6X/ 1 .

Reference

s Adzhubei , I. , Jor d a n, D .M. , Sun yaev , S.R . (2 013) Predic ti ng func t i onal e ff ec t o f hu man missen se mutation s u sing Po lyP he n-2 . Cur r . Pr ot oc . Hum. Ge net . 7 Unit7 .20 . Bet ter s , J.L ., Yu , L. (20 10) N PC1L1 and c h ole ste r ol t r a n s por t. FE BS L e t t . Chan g, C .C ., C how, C .C . , Tell ie r, L.C . , V at ti kuti, S ., Purcel l , S.M. , Lee, J . J. (2015 ) Se c ond-ge n era tion PL IN K: r i sing t o the ch all eng e o f large r a n d riche r da ta s e t s . Giga s c ie nce 4, 7. Cuc hel, M ., Br uc ke rt, E ., G in sb erg, H . N., Ra al, F . J . , S anto s , R. D., Heg e le, R .A ., K uivenho ven, J .A ., Norde stga ar d, B. G., De scamp s, O.S ., S te i nhage n-Thi e s s e n, E. , T ybjærg-Han sen , A. , W a tt s, G .F., Averna , M ., Boil e au, C., Bo r é n , J. , Ca tapa no, A.L. , De fe s c h e, J .C ., H ov ingh, G .K., H u mp hr i e s , S.E. , Kov anen, P.T ., Ma sa na, L., Pajuka nt a , P., P a r ho f e r , K. G ., R ay, K.K., Sta l enho ef , A.F . H ., Str oe s , E., Taski nen , M.R ., Wi e gman, A. , W i klund, O., C ha pman, M . J. ( 2 014) Homozygou s f a mili al hyp er c hole s terol aemi a : New in sig hts an d guidan ce for c linic ia ns to improve d ete ction and c linic al mana gement . A po s i tion pap er f romth e Con sen su s Panel o n Famil ial Hype r c hol e s terol aemia o f th e Europea n Athe ro s c ler o s is S o ciety. Eur . H eart J. Curti s, D . (2012) A r apid met hod f or com b ined an aly si s of comm on and rar e v aria nts at th e lev el o f a region , gen e, or pa thwa y. Adv A p pl Bioi n form Ch e m 5, 1– 9. Curti s, D . (2016) Pa t hw ay analy s i s o f wh ole ex ome s equenc e da ta provide s f urth er s uppo rt f or th e inv olveme nt of hi s t on e modi fic a tion in the aeti olog y of schi z oph reni a . Psych iat r . G e ne t. 26, 22 3–7 . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint Curti s, D . (2020) An al ys i s o f e xome - sequ enc ed UK Bioba nk subjec ts imp lica t e s g e nes a f fect ing r i s k o f hyp er l ipida em ia. Mol . Gene t. Me tab . Curti s, D . (2021) Mul tiple Line ar Regr e s s i on Allows Weigh ted Burd en Analy s i s o f R are C oding Vari ant s in a n Ethnic ally H e t e rog en eou s Popula t io n . Hum. Her ed. 1–1 0 . Doggrel l, S .A. (20 20) Will ev inac umab be come t he s ta nd ard t re a t me n t f or homozy gous famili a l hyp er c hole s terol emia ? Ex per t Opin . Biol. The r . 1–4. Dron, J.S ., Heg el e, R. A. (2016) G ene tic s o f L ipid and L ipopro tein Dis or d e rs and Tra its. Cu r r. Gen et. Med. R e p. 4, 13 0–141 . Eng elen, M. , Kemp, S ., De Vis ser, M ., Van Geel, B. M. , Wand er s , R .J .A. , Aubou rg, P. , Poll-Th e , B.T . (2012 ) X-linke d adre noleu kody st r op hy (X -ALD ): Clin ical pre sent atio n and g uidelin e s for dia gn osi s, follow -up and manage m ent . Or ph a ne t J. Rare Dis . Galin sky , K.J. , Bha t ia , G ., Lo h, P.R ., G eorg ie v, S ., Mukhe rjee , S ., P a tte rson , N. J., Pri c e, A.L. (2016) Fa s t Princ ipal - Compon en t Ana lysi s Reve al s Co nverge nt Evolutio n of A D H1B i n Euro pe a nd East A s ia . Am . J. Hum. G ene t. 9 8, 456– 472 . Gr e t a rsd otti r, S. , Helg a son, H ., Helga d ott ir, A., Sigur ds son, A ., Thorlei f s son, G., Ma gnusdo ttir , A. , Odd s s on, A ., St einth or s do ttir , V. , Ra f n ar, T., de G r aa f , J ., Da ne shpou r, M. S., H e d ay ati, M . , Az izi, F ., Gr a ru p, N. , Jø r g en sen , T., Ve s t ergaa rd, H., H a n s e n , T., E yjolf sson, G., Sigu r da r do tti r, O., Ola f s s on, I . , Kiem eney, L.A. , Pe der sen, O., Sulem, P ., Thorge irs son, G., Gudbjar t s s on, D .F. , H o l m, H., Thors t e in s d ot ti r, U. , S t e f a n sso n, K. (2015 ) A Sp lice Region Va r i ant in L DLR L ower s N on- hi gh Den si ty Li poprote in Chol e s t erol an d Pro t e ct s aga in s t Cor onary A r te ry Di sea s e. PL oS Gen et . 1 1. Kanu r i, B. , Fong , V ., Hall e r , A ., Hui, D.Y., Pat el, S . B. (2020 ) Mic e la cki ng glo bal S t a p1 exp r e ssi on do not man if e s t hy pe rchole sterol emia . B MC Med. G ene t. 21. Kishnani , P.S ., Au stin, S .L. , A b d enur, J.E., Arn, P ., Bali , D . S ., B oney , A ., Chung , W.K., Dagl i, A. I., Dal e , D., Koe b er l , D. , Some rs, M .J ., Bur ns Wec hsle r, S. , Wein stei n, D. A., W olf s d o r f , J. I., W a tson , M.S . (2014 ) D i a gno si s and man ageme n t o f glyc ogen storage di se a s e type I: A p r a ctice g uide line o f th e Americ an C olle ge o f Me di cal G e netic s an d Geno mic s. Gene t. M ed. 1 6, 1 –29 . Kolda mova, R., Fi tz, N .F., L e ft erov, I . (201 4) ATP-bin ding c a sse tte tran sp or t er A1 : F r o m metaboli s m to neur odeg en era tion . Neu robiol . Di s . Kuma r , P. , Heniko ff, S ., Ng, P. C . (2009 ) P r edi cting the ef fec t s of codi ng non - synon y mous varian ts o n prote in func tion u sing the S IFT algorit hm. Nat . Protoc . 4, 107 3–1081. La Rosa , J. C. , He, J., Vu ppu t u r i, S . (199 9) E ff ect o f s ta tin s on ri sk of c oron ary di sea s e. A me ta -a naly si s of rand omi z ed con troll ed tri al s. J . Am. Me d. A ssoc . Lu , Z., L uo, Z ., Jia, A ., Yu, L ., Muh amma d, I., Z eng, W ., S ong, Y . (2018 ) A s s o cia tion s of th e AB CA1 gene poly morphis m s w ith pla sma l ipi d lev el s. Medic ine ( Bal t i m or e ). 97, e1 3521 . Ma, Y ., Han , X., Zhou, X ., Li, Y., G ong, S., Z hang, S ., Cai , X., Z hou, L. , Luo, Y., L i, M ., L iu, W., Z hang, X., Ren, Q . , Ji, L. (20 19 ) A new cli nical s c re en ing s tr at egy and prev alenc e e stima tion f or glu coki nase va r i ant -induce d di abe te s in an adu l t Chin es e popul a tion . Gen et . Med . 21, 939– 94 7. Marang hi, M ., Trugl io, G., Gallo, A., Gr ie c o, E ., V erri enti , A., M onta li , A. , Gall o, P., Ale sini, F. , Arca , M., Lu carell i, M. (2019 ) A novel s pli cing muta tion in t h e AB CA1 gene , cau sing Tang ier dise a se and famil ial H DL de ficie n cy i n a la r g e famil y. Bioc hem. Bi op hys . Re s. Commun. 508, 4 8 7–493. McL aren, W ., Gil , L., Hunt , S .E., Ri a t , H.S ., Ritc hie, G .R. S., Thorm ann, A., Flic ek, P., Cunnin gham, F . (2016 ) The En sembl V arian t Ef fec t Pre dic tor. Gen ome Bi ol. 17, 122 . Nata rajan, P., Pel o s o , G. M., Z ekav a t, S.M ., Mon t a sser , M., Ganna , A., Cha ff in, M . , Khera, A . V ., Zhou, W., Blo om, J. M., Engrei tz, J.M ., Ern st, J. , O’C onnell, J . R ., Ruo t s a lain en, S.E ., Alver , M., Ma nich a ikul, A., J ohn son, W .C ., Perr y , J.A ., P ot erba, T . , See d, C ., S ura kka, I.L. , Esko , T., Ripa tti, S . , Salo maa, V., Corr ea, A ., Va s a n, R .S. , Kelli s, M., Neal e, B.M. , L ander, E . S., A bec a si s , G., M i t c hell, B., Ric h, S.S ., All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint Wils on, J .G ., Cup ple s , L .A ., Ro tt er, J .I ., Wi lle r , C .J ., Kathi re san , S ., Ab e, N ., Alb e r t , C ., Allr ed, N . (Nichol e ) P., Alma s y, L., Al on s o, A. , Ame n t, S., An der s o n, P ., Anugu, P ., Appleb a um-Bowden , D. , Arking, D. , Ar n e tt, D .K., A s hl e y-Koch , A ., As l ib eky an, S ., Assim e s, T., Au er, P., Avr a mop oulo s , D. , Barnar d, J ., Bar ne s , K., B arr, R .G ., Ba rron - Ca sella , E., Be a t y , T., B eck er , D . , Beck er, L., Bee r, R., B e gum, F., Bei tel sh ee s, A ., Benj amin , E ., Bez erra , M., Biel a k, L ., Bi s, J ., Bla ckw ell, T. , Bla nge ro, J. , Boe rw inkle, E., Bor e cki , I., B ow ler, R ., Brody , J., B roec k el, U., Br oome, J . , B unting, K., B urcha r d , E ., Cardw el l, J. , Car ty, C . , Ca sa buri, R ., C ase lla, J., Chang, C ., Cha sman, D., Chava n, S ., C h en, B .J ., C he n, W. M., Chen , Y.D. I., Cho, M ., Ch oi , S.H ., Ch ua ng, L .M ., Chung , M. , Co rnell, E. , Cr anda ll, C., Cr a p o , J., Curr an, J ., Curti s, J. , Cu ster, B . , Damcot t, C ., Da r ba r , D., Da s , S ., Davi d, S. , D a vi s, C ., Day a, M. , de Andr a de, M ., DeBa un, M. , Dek a, R., DeM eo, D., Dev ine , S ., Do , R., Duan, Q., Dug girala, R ., Du rda , P., D utc h er, S. , Eaton, C., Ek unwe, L., E llino r, P., Emery, L ., Fa r b e r , C., Fa r na m, L. , Fing e r l in, T., Flic king er , M., Forna ge, M ., F ra nce schi ni, N ., Fu, M ., Ful l ert o n, M., Ful ton , L., Gab riel , S ., Gan , W. , Gao, Y. , G a ss, M ., Gelb, B . , Ge ng, X. (Pr i sci lla ), G erme r, S. , G i gnoux, C. , Gladw in, M ., Gl ah n, D., Gog arte n, S ., Gong, D.W ., G or i ng, H. , Gu, C. C. , G u an, Y. , G uo , X., Hae s s l er, J. , Hall, M ., H a rri s, D ., Hawl e y, N., He, J., Heav ner, B ., Heck b ert , S., He rnand e z , R. , Her ring to n, D ., He r s h, C . , Hidalgo , B., Hix son, J., Hok an son, J., Hong, E. , Hoth , K., H siung, C . (Agn e s), Huston, H. , Hw u, C. M., Irvin , M.R ., J ac kso n, R., J ai n, D ., Jaqui sh, C., Jhun, M .A. , John s en, J ., Jo hn s on, A., J o hn s t on, R. , Jon e s , K., Ka ng, H. M. , Kapl an, R., Kardia , S. , K aufman , L ., Ke lly, S ., Kenn y, E ., Ke s sl er, M ., Khan, A., Kinney, G ., Konkl e , B., Koop e r b e r g , C., K r a m er , H . , Kraut er, S ., L ang e, C., L an ge , E., L ange , L., Lauri e, C athy, L auri e, C e cel ia, L eBo ff , M., Le e, S. S ., L ee, W.J ., L eFai ve , J. , L evi ne, D ., L evy, D . , Lew is, J., Li , Y., L in, H. , L in, K. H ., Liu, S ., L iu, Y., Lo os , R. , Lubi t z , S. , Lune t ta , K., L uo, J ., M aha ney, M ., M a ke , B. , M an son, J .A ., Mar goli n, L., Martin , L., Mathai , S., Ma thia s , R ., McA rdle, P. , McD onald , M.L ., M c Farl a nd, S. , M c Ga rvey, S., Mei , H . , M eyer s, D.A ., Miku lla , J. , Mi n, N. , Mine ar, M ., M i n s t er, R .L ., Mu sa ni, S ., Mwa s ongw e, S. , Mycha lec kyj, J .C . , Nadka r ni , G. , N a ik, R . , N e kh ai, S., Nic ker son, D. , N o rth, K. , O ’ Co nno r , T., Och s -Balc om, H., Pa n kow, J . , Pap anico l aou, G., Pa r k e r, M. , Pa r sa, A ., P e nchev , S., Per alta , J. M., P ere z , M., P ete r s , U ., P eys er, P ., P hil lip s, L., P hillip s, S . , P olli n , T., Po st, W ., Bec ker, J .P., Boo rgula, M. P., Preu s s , M ., P r o kop enk o, D ., Ps aty, B ., Q a sba , P. , Qiao, D., Qin , Z ., Ra f ae l s , N., R a ffi eld, L ., R ao, D. C ., Ra smu sse n-Torvi k, L ., Ra tan, A., Re d line, S ., Re ed, R. , Reg an, E ., Rein er , A., Ri ce , K ., Rod en, D., Ro s el li, C ., Ruczin s k i, I ., Ru ssell, P., Ruuska , S., Ry an, K. , Sak or n sak olp at, P., S alimi , S., Salzb erg, S . , Sa ndow, K. , S ank ar an, V ., Schmi dt, E. , Sc hwan der, K., S chwa rtz, D., Sc iur ba, F . , Se idman, C . , She e han , V ., S he tty, Amol, S hetty, Anik et , She u, W .H . H., Shoema k er, M .B. , S ilver, B . , Silv erman, E . , Smith, Je nni fer , Smith, Jo sh, S mit h, N. , Smi t h, T., S molle r, S ., Sniv ely, B ., So fe r, T . , Sotood eh nia , N., S t i lp, A ., S tree ten , E., S u ng, Y.J., Sylv ia, J ., Szpiro, A ., S z t alry d, C ., Ta liun, D. , T ang, H ., Taub, M. , Tayl or , K., Ta ylor , S. , Tel en , M., Thorn ton, T . A . , Tink er, L. , Tir s c hwe ll, D ., Ti wari, H. , T r a cy, R., Ts ai, M. , Vai dya , D ., V andeHa a r , P., Vrie ze, S. , W a lke r, T., W a llac e , R., Wal ts, A., W an, E., Wang, F.F. , Wa tson , K ., We eks , D .E., Wei r , B ., W eis s , S ., W eng, L. C., Willer , C. , W i lliam s, K., Wi lliam s, L .K. , Wils on, C ., W ong , Q. , Xu, H. , Ya nek, L., Y ang, I. , Yang , R ., Za ghloul, N ., Zhang , Y., Z hao , S .X., Zhe ng, X., Zhi, D., Zhou, X. , Zody , M. , Zoe llne r, S . ( 2 018) Deep - co verage whole genom e seq uenc es an d blood l ipid s among 16,324 indivi dual s. N at. Commun. 9, 1– 12. Punto ni, M. , Sbrana , F., Bi ga zz i, F., S ampi etro , T. ( 2 012) Ta ngi er Di s ea se . Am. J . Ca r di ovasc . Drug s 12, 303– 311. R Core T e am (201 4) R : A lan guage an d e n vironment for sta t i s tic a l co mputing. R F oundati on fo r Stati s tica l Compu ting, Vienna , Au s t ria. , A ust r i a. Sha r i fi , M., Fu tem a , M., Nair, D., Humph r ies , S. E. (2017 ) Gene t i c Archit ec ture o f Fam ilial Hyperc hole ste r ol a emia . Curr. Ca r d iol . Re p. Si ngh, T., Kurki, M., Curtis , D ., Pu rc ell, S . , et al . (2016) Ra r e SETD1 A lo ss -o f-func tio n va r i ant s a re as s o cia ted with schizo phre nia and de vel opme nt a l di so rde rs. Na t N euro s c i. Sub r a manian , A. , Tama yo, P. , Moo t h a, V. K., Muk herje e, S ., Eb e r t , B.L ., Gille tt e, M. A., P aulov ich, A . , Pome roy , S.L ., G olub, T .R ., La nde r, E.S ., Me sir ov , J .P . (20 05) Gen e s et e nrichm en t analy s i s: a kno wledg e-b as ed app roac h f or int erpr eti ng ge nome-wi de ex pre s sion pro file s . Pr o c N a t l Acad S ci U S All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint A 102 , 15545– 15550. Szus t a kow ski, J. D . , Bala sub ramani an , S. , Sas s o n , A. , Khal id, S. , B r o n s on , P. G . , Kvik s t ad, E., W ong, E., Li u, D., Davi s, J. W., Ha e fl iger, C., Loomi s, A.K., M i kkil inen i, R. , N o h, H .J . , Wadh aw a n, S., Bai, X . , Hawe s, A., Kra she ninin a, O . , Ulloa , R., L o pez, A. , Smi t h , E. N., Waring, J., Whe lan , C. D., T s a i, E. A., Overto n, J ., S ale rno, W ., J ac ob, H. , S z al m a, S., Runz, H . , H i nkle , G . , N io i, P., Pe t rov ski , S., Mi ller , M.R ., Bara s, A ., Mi tnau l, L. , Reid, J. G. (202 0 ) Ad vanc ing Huma n G e ne tics R e sea rch and D rug Disc ov er y through E xome S eque nc ing of t h e U K Bio bank . medRx iv 20 20.11.02 .2022223 2. Wang, X., Mu sunu ru, K. (201 9 ) Angiopoi e t i n -Like 3: From D i sco ve ry to Th er a peu t ic Gene Editing . JAC C Ba sic to Tran s l . Sc i. Whyte, M .P. , Aron son, J., McAli s te r , W . H., W e in stein , R.S. , Wenk ert, D . , Clem ent s, K.L ., Got t e s man , G.S. , Ma d s on , K.L., St olina, M. , Bijank i, V . N., Plotk in, H ., Hu s k ey, M . , Duan , S. , Mu mm, S. (2020) Coal e s c ing E xpan sile Ske l etal Di sea s e: D e linea ti on O f A n Extraor dinary Os teop ath y I nvolvi ng The IFI TM5 Mut atio n Of Ost eogene s i s Imp er f ec t a Type V . Bone 145, 11 5835. Willer, C. J ., Sc hmid t, E.M ., S engup ta, S. , Pe lo s o , G . M., G u sta f sson, S ., Kano ni, S ., G a nna , A., Ch en , J. , Buc hkovi ch, M.L ., Mor a, S . , Bec kmann, J. S ., Bragg -G r e sham, J .L ., Chang , H.Y ., D emirk an, A. , Den Hert o g, H .M. , Do , R., Donne lly, L. A., Eh re t, G. B., E s k o, T., F eito s a, M.F ., Fe rrei ra, T. , Fische r, K ., Fon t a nill as , P., Fra se r , R. M., F reit ag , D .F., Gurda sani , D. , Hei kkilä , K., Hyp pön en, E. , Isa ac s, A. , Jac kso n, A. U . , Johan s s on , Å., John s o n , T., Ka akin en, M., Ke t tu nen, J., Kleb e r , M.E ., Li , X., L uan, J., Ly ytikä inen, L .P. , Ma gnu sson, P .K.E. , Ma n gino, M., M i h ailov , E., Mon ta s se r, M.E ., Mülle r-Nu ra syid, M., Nol te , I. M., O’Con nell, J.R ., P a lmer, C .D. , Pe rola , M., Pete rs en, A.K . , Sann a , S., Sax ena, R. , S ervic e, S.K ., Sha h , S., Shungin , D., Sid ore , C. , Son g , C., S traw bridge , R. J . , Sur akk a, I. , Tana k a, T. , Te slovi ch, T.M., Thorl eif s s o n, G., Va n De n Herik, E. G . , Voight , B.F . , Volc ik, K . A . , Wait e, L .L., Wong , A. , Wu, Y., Zha ng, W., Ab sh er, D., A siki, G., B arros o, I., B een, L .F ., Bo l t o n, J.L., B onnyc astle , L. L ., Brambill a, P. , Burne tt, M .S. , C e sa na, G ., Dimitrio u, M., Done y, A .S. F ., D öring, A ., Elliot t, P., Ep ste i n, S.E ., Eyj olf sson, G.I ., Giga nt e , B., Good arzi, M .O ., G r a lle rt, H . , Grav ito, M .L. , Grov e s , C.J ., Hallma n s, G., Har tika in en, A.L. , Haywa r d , C. , Hern andez , D. , Hic ks, A .A., H olm, H., Hung, Y. J., Ill ig, T. , Jo ne s , M .R., Kal e ebu, P., Kast elein , J. J . P., Khaw , K.T. , Kim, E ., Klop p, N ., Komulain e n, P., K umari, M., L ange n berg, C ., L ehtim äk i, T., Lin, S.Y., Lind ström, J. , Loo s, R. J.F ., Ma ch , F., Mc Ar dl e, W .L., Mei s i nger, C., Mi tc hell, B . D . , Müller , G., N agaraj a, R ., N ari su, N ., Ni emine n , T. V.M ., N s u buga , R. N ., Olaf s so n, I. , Ong, K . K., Pal o t i e, A., Pap ama rkou, T., Pomil la, C., Pout a, A ., R a der, D. J . , R eilly , M. P., R idk er, P.M ., Riva de neir a, F. , Ru dan, I., R uokon en, A. , Sama ni , N. , Sc harnag l, H ., Se el ey, J . , Sila nde r, K., S t a ncá ková, A ., Stirrup s, K., Sw ift , A.J ., Ti r e t, L ., Ui tt erlinde n, A. G. , Va n Pel t, L .J., Vedan t a m, S ., W a inwrigh t, N ., Wijme nga, C . , Wild, S.H., W il l em sen, G., Wilsg aard, T. , Wil son , J.F. , Young, E.H. , Zha o, J .H . , A dair, L .S ., Arvei ler , D. , Assime s , T. L., Ba ndi nell i, S ., Benn et t, F. , Bochud, M . , Boehm, B .O ., Boo m s ma , D . I. , Boreck i, I .B. , Born stei n, S.R ., Bove t, P., Bu rnier , M., Ca mp bell, H., Ch akrava r ti, A. , Chambe rs , J. C., Che n, Y .D .I. , Coll in s , F. S., Coop er, R. S., D a ne sh , J. , De d ous s i s, G., De F ai re, U ., F e r a nil, A .B. , Fer ri ère s , J ., F errucc i, L. , Freime r, N. B., Gie ge r, C ., G roop, L . C. , Gudna son , V. , Gy llen s t e n, U ., Ham sten , A., Har r i s , T .B ., Hingorani , A ., Hir s c hho rn, J .N. , H o f ma n, A., Hovi ngh, G .K., H siung, C.A ., Humphri e s, S.E ., H un t , S. C., Hveem , K., Irib a r r en, C . , J ärvelin , M .R. , Ju la, A., Käh öne n, M ., Kap r i o, J ., Ke säniemi , A., Kiv imaki , M., Koone r , J.S . , Koud sta al, P.J ., Krau s s , R. M . , Kuh, D., Ku u s i st o, J ., Ky vik, K.O ., Laa ks o, M., L akka , T.A. , Li nd, L. , Lin dgren, C .M. , Ma rtin, N.G ., Mä rz, W . , McCa r thy , M .I. , McK enz i e , C. A., M ene ton, P., Met s p al u, A. , Moi lan en, L., Mo rri s , A .D., Munr o e , P. B., Njøls tad , I. , Ped er sen , N.L. , Power, C., Pram st aller , P. P . , Pric e, J .F., P saty, B .M. , Q u e r t ermou s, T. , Raur amaa , R ., Sa leh een , D., Sal omaa, V., Sa nghera , D.K ., Sarami e s , J ., Schw arz , P.E .H., Sh e u, W .H.H . , Shuldi ne r , A .R., Si e gba hn, A., Sp ec to r , T.D., S te fan sson , K. , Str acha n, D. P., Ta yo, B.O. , T r e moli, E. , Tuomile hto , J. , Uu s itup a , M., Van Duij n, C.M ., V ol lenwe id er, P., Wall en tin, L ., Wa reham, N .J ., Whi tfi eld , J .B ., Wol f fenbu tt e l, B.H.R ., O rdova s, J.M ., Boe rw inkle, E., Pa lme r, C .N. A., Tho r ste in s do ttir , U., Cha sman , D . I., Ro tt er, J.I . , Frank s, P .W . , Ripa tti, S., Cuppl e s, L.A. , S andhu, M .S . , Ri c h, S.S ., B oehnk e , M., Delo uka s , P. , Ka thi r e s an , S ., Mohl ke , K.L ., Inge l sson, E . , Abeca s i s , G. R. (201 3 ) Di s c overy and re fin eme nt o f loci a ssoc ia t ed wi t h lipid l evel s. Nat . Gen et . 45, 1274–1 285. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint Wu, Y., Byrne , E. M . , Zheng , Z., K emper , K.E. , Yeng o, L., M alle tt, A.J ., Ya ng, J ., Vis sc her, P.M ., W r ay , N.R. (2019) G en om e-wide a ss oc iation s tu dy of me dicati on - u s e a nd a s s oc ia ted d i sea s e in th e UK Bioba nk. N at . Commu n. 10, 1 – 10. Yan g, R., L iu, N. , Ch e n, L. , Ji ang, Y., S hi, Y. , M a o, C ., L iu, Y ., W ang, M. , Lai, W . , Tang , H., Ga o, M. , Xi ao, D., W ang, X., Zh ou, H. , Tang , C. e. , Liu , W. , Yu, F ., Cao, Y. , Ya n, Q., Liu , S ., T ao, Y. (2 019) GI AT4RA func t i o ns a s a tumor s uppr e s s o r i n non - sma ll c ell lu ng c ancer by cou nt e ra cting Uc hl3–medi a ted deubi quiti nati on of LSH. Oncog ene 38 , 7 133– 7145. Zha ng, Y. , Gu, J ., Wa ng, L., Z hao, Z ., Pan, Y., Chen , Y. (201 7) Ablati on of PPP 1R3 G r educe s g lyco gen depo sitio n a nd mitiga te s hi gh- fa t diet in d uced o be s i t y . Mol. Cel l . Endoc rinol . 43 9, 133– 140. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint Tabl e 1 The t a ble show s th e w eight w hich was a s s i gned to e ach ty p e of va r i ant a s ann ota t ed b y VEP, Poly phen a nd S IF T a s we ll as the br oad c atego r i e s w hich were u sed for multiv a ria te ana ly se s o f va r i ant e f fect s ( Adz hub e i et al ., 2013; Ku mar et al . , 2009; M c La r e n et al ., 2016) . VEP / SIFT / Pol yph en anno t ation Weig ht Ca t e gory interge ni c_v arian t 1 Unus ed fea tur e_t runca t i on 3 In t ro nic , etc . regula to r y _re gion_v ari ant 3 In t ro nic , etc . fea tur e_e l ongati on 3 In t ro nic , etc . regula to r y _re gion_a mpli fic a tion 3 In t ro nic , etc . regula to r y _re gion_a blati on 3 In t ro nic , etc . TF _binding _ s i te _vari ant 3 In t ro nic , etc . TF BS_ampl ific a t io n 3 In t ro nic , etc . TF BS_abl a tion 3 In t ro nic , etc . dow ns t ream _g ene _v arian t 3 In t ro nic , etc . ups tr e a m_gene _varia n t 3 In t ro nic , etc . non_c oding _ tr a n s c r ip t_v arian t 3 In t ro nic , etc . NMD_tr an script _v arian t 3 In t ro nic , etc . intron _va r i an t 3 In t ro nic , etc . non_c oding _ tr a n s c r ip t_e xon_v ari ant 3 In t ro nic , etc . 3 _ p r i m e_U TR _ v a r ian t 1 0 3 pr i me UTR 5 _ p r i m e_U TR _ v a r ian t 5 5 pr i me UTR mature _mi RNA_v aria nt 5 Unus ed co ding_s equen c e_v ari ant 5 Unus ed synony mou s _ vari ant 5 S ynony mous stop _re ta ined _v arian t 5 Unus ed inc omplete _t ermina l _cod on _varia n t 5 Unus ed splice _regio n_v aria nt 5 S plic e region prote in_a l t e r i ng_va r i an t 10 Pr ot e i n a lt e r i n g m i ss ens e_v a r i a nt 1 0 Pr ot e i n a lt e r i n g infram e_ de leti on 15 In D e l , etc infram e_ in s e r tio n 15 In D e l , etc tran s c rip t_ampli fi ca tion 15 In D e l , etc sta r t _lo s t 30 Unus ed stop _lo st 30 Unus ed frame shi f t_varia n t 40 Di srupti ve s t o p_ g a i ned 4 0 Di srupti ve splice _dono r_va r i a nt 40 Sp l i ce s i t e v a r i a n t splice _ac cep tor _va r ia n t 40 Sp l i ce s i t e v a r i a n t t r a ns cr i pt _ ab l at ion 2 0 Di srupti ve SIF T del e ter i o us 10 Dele t e r io u s Poly Ph en po s sibly d amagi ng 10 Po ssibly damagi ng Poly Ph en proba bly da maging 10 Proba bly da maging All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint Tabl e 2 Gene s with a b s ol ute val ue o f SL P ex cee di ng 3 or m or e (equiva l ent to p<0 . 001) f or te s t o f a ssocia t i on of w eight ed b ur de n s c o re with hyperl ipi da emia. Sy mbol SL P N a me LDL R 50.08 Low Densi t y Li poprote in Rec epto r G6PC 5.55 Gluco se -6- Pho sph ata s e Ca tal ytic S ubuni t SULT 1E1 4.63 Sulfo tran s fer a s e Fami ly 1E Member 1 LOC101 928415 4.50 Uncharac t er i zed LOC101928 415 SL C35G1 4.38 Solute Car r i er Famil y 35 Me mbe r G 1 PL A2G 5 4.15 Pho spholip a se A2 Group V CMTM7 3.99 CKLF Li ke MARVEL Tran smembra ne D oma in Con taini ng 7 MIR6716 3.95 MicroRN A 6716 COL4A2 - A S2 3.85 COL4A2 An ti sen se 2 DEFB13 1A 3.66 De fen sin B eta 13 1A OTULIN 3.62 O T U Deu bi quitin as e Wi th L inear L inkag e Spe cific ity FAM122 C 3.51 Fami ly With S equ enc e Simi lari ty 122 C CMIP 3.48 C- Maf Induc ing Protei n EIF 4B 3.45 Euka r y otic T r a n s la t i on Ini tiati on Fac t o r 4B PP P2 R3B 3.41 Pro tein P ho spha ta se 2 Regul a t o ry Subunit B'' Beta HNR N P A B 3.38 Hete r o gen eou s Nucle ar Ribonuc l eopr ote in A/B PREB 3.37 Prolac t i n Regul a tor y Eleme n t Binding PEX1 2 3.36 Peroxi soma l Biog en e s i s F act or 12 FAM167 A 3.35 Fami ly With S equ enc e Simi lari ty 167 Me mber A ABCG5 3.31 ATP Bin ding C a ss et t e Sub fami ly G Memb er 5 ABCD1 3.26 ATP Bin ding C a ss et t e Sub fami ly D Memb e r 1 PRAF2 3.21 PRA1 D omain Fami ly Membe r 2 CTHRC1 3.16 Colla gen Tripl e Helix Rep eat Con ta ining 1 SL C25A37 3.14 Solute Car r i er Famil y 25 Me mbe r 37 CT62 3.11 Canc er / Te sti s A ssocia te d 62 L1 TD1 3.09 LINE1 Type Tr a ns p o sa se D omai n Co ntain ing 1 PIK3R6 3.09 Pho sphoin osi tide -3 -Kina s e Re gulat ory S ubunit 6 FOX O3B 3.08 Forkhe ad B ox O 3B FAM47A 3.06 Fami ly With S equ enc e Simi lari ty 47 Member A MIR6806 3.06 MicroRN A 6806 GCK 3.04 Glucok ina se MAPKA PK2 3.02 MAPK Activ a ted Pro tein Ki na s e 2 HLA- A 3.01 Major His t oc ompa tibil i ty Comple x, Cl a ss I, A CRYZ L1 -3.06 Crys t a llin Z e ta Li ke 1 A PP BP2 -3.08 Amyloi d Beta Prec u rsor P rot ein Bind ing Pro tein 2 GFPT1 -3.10 Glutamin e- -F ruc t o se -6 -Pho spha te T r a n s a m inas e 1 TTR - 3 . 1 6 T r a n st hy r et i n TXNL4 A -3.22 Thioredox in L ike 4A ITM2B -3.23 Integr al M e mbra ne Pro tein 2 B All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint UBR4 -3.23 Ubiquitin Pro tein Liga se E3 Compon e nt N -R ec ognin 4 SV2B -3.29 Syn aptic Ve s ic l e Glyc opro tein 2B LOC101 929609 -3.30 Uncharac t er i zed LOC101929 609 LOC105 377994 -3.40 Uncharac t er i zed LOC105377 994 SNX17 -3.62 Sorting N ex in 17 ANGPT L4 -3.66 Angiopoi e tin Li ke 4 NPC1L1 -3.70 NPC1 Li ke Intr ac ellula r C hol e st e r ol Tran sp or te r 1 CTXN2 -3.91 Cor t e xin 2 TBC1D8 -3.93 TBC1 Doma in F a mily Mem ber 8 LOC107 985474 -3.95 Uncharac t er i zed LOC107985 474 PP P1 R3G -4.25 Pro tein P ho spha ta se 1 Regul a t o ry Subunit 3G A PO C 3 -4.89 Apolipopr ot ein C3 A NGPT L3 -5.67 Angiopoi e tin Li ke 3 LOC102 723729 -5.77 Uncharac t er i zed LOC102723 729 IFI TM5 -5.86 Inte rf eron Induc ed Tran sm embran e Pr o t ei n 5 PCS K9 -10.42 Propr ote in Conv e r ta s e S ubtili sin / K exin Type 9 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint Tabl e 3 Res ult s from l ogis t i c regr e s s i on ana ly si s s howi ng the c ontri bution di f fer ent c a tego rie s of va r i a nt wi t hi n a g ene mak e t o ri sk of hy perli pida e mia. Odd s r atio s for e ach c atego ry are e stima ted inc ludi ng princ ipal c omponen t s and sex a s cov ariat es . Tabl e 3A Res ult s fo r LD L R . Category Total cou nt in con t rols Mean c ount i n c ont rols Total count in cases Mean count in ca s es O R (95 % CI) SL P Intro nic, etc 2720 7 0. 173760 7854 0. 178281 1.01 (0.99 - 1.0 4) 0.47 5 prime U T R 5 5 0. 000351 21 0. 000477 1.43 (0.85 - 2.4 1) 0.78 Sy nonymous 258 2 0. 016490 682 0. 015481 0.92 (0.84 - 1.0 0) -1.31 Spl ice r eg ion 664 9 0. 042464 1735 0. 039383 0.86 (0.81 - 0.9 2) -5.21 3 prime U T R 50 6 0. 003232 139 0. 003155 0.98 (0.81 - 1.1 9) -0.06 Pro tein a l t e r i ng 494 7 0. 031594 1800 0. 040859 1.15 (1.05 - 1.2 5) 2.87 InDe l, etc 0 0. 000000 10 0. 000227 6.58 Di sr upt iv e 3 0. 000019 31 0. 000704 40 .02 (11.8 3 - 135 .3 3) 1 6.95 Spl ice s i te v arian t 2 0. 000013 11 0. 000250 23.31 (4. 96 - 109 .4 2) 5.55 Dele te riou s 70 2 0. 004483 473 0. 010737 1.74 (1.41 - 2.1 4) 6.87 Po s s i bly da maging 40 0 0. 002555 200 0. 004540 1.20 (0.96 - 1.4 9) 1.02 Proba bl y dama ging 53 9 0. 003442 350 0. 007945 1.30 (1.03 - 1.6 5) 1.66 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint Tabl e 3B Res ult s fo r PC S K 9 . Category Total cou nt in con t rols Mean c ount i n c ont rols Total count in cases Mean count in ca s es O R (95 % CI) SL P Intro nic, etc 531 7 0. 033958 1540 0. 034957 1.02 (0.96 - 1.0 8) 0.31 5 prime U T R 26 9 0. 001718 78 0. 001771 1.03 (0.80 - 1.3 4) 0.10 Sy nonymous 723 0 0. 046175 2145 0. 048690 1.02 (0.97 - 1.0 7) 0.40 Spl ice r eg ion 29 1 0. 001858 90 0. 002043 1.04 (0.81 - 1.3 3) 0.12 3 prime U T R 241 8 0. 015443 707 0. 016048 1.02 (0.94 - 1.1 2) 0.24 Pro tein a l t e r i ng 338 8 0. 021638 831 0. 018863 0.96 (0.86 - 1.0 6) -0.44 InDe l, etc 4 0. 000026 3 0. 000068 1.81 (0.39 - 8.4 8) 0.74 Di sr upt iv e 20 2 0. 001290 29 0. 000658 0.51 (0.34 - 0.7 7) -3.05 Spl ice s i te v arian t 17 9 0. 001143 14 0. 000318 0.29 (0.17 - 0.5 1) -5.08 Dele te riou s 140 4 0. 008967 279 0. 006333 0.59 (0.45 - 0.7 7) -4.00 Po s s i bly da maging 24 9 0. 001590 68 0. 001544 1.19 (0.88 - 1.6 1) -0.60 Proba bl y dama ging 105 0 0. 006706 227 0. 005153 1.29 (0.96 - 1.7 4) -1.10 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint Figure 1 QQ plot of SLPs obtained for weighted burden analysis of association with hyperlipidaemia showing observed against expected SLP for each gene, omitting results for LDLR and PCSK9 . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted January 27, 2021. ; https://doi.org/10.1101/2021.01.05.20249090doi: medRxiv preprint

Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.

My notes (saved in your browser only)

Ask this paper AI returns verbatim quotes from the full text · source: oa-pdf

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Citation neighborhood (sparse)

Too few in-corpus citations on either side for a chart; here are the lists.

Cited by (1)

Cited by (1)

Source provenance

europepmc
last seen: 2026-05-19T01:45:01.086888+00:00
unpaywall
last seen: 2026-07-19T06:49:21.617583+00:00