Genome-wide characterization of copy number variants in epilepsy - - PowerPoint PPT Presentation

Genome-wide characterization of copy number variants in epilepsy patients

Canadian Human and Statistical Genetics Meeting

Jean Monlong April 24, 2017

BOURQUE LAB MCGILL UNIVERSITY HUMAN GENETICS DEPT.

2

Copy Number Variation (CNV)

Imbalanced genetic variation involving more than 500bp.

3

Epilepsy

Neurological disorder characterized by recurrent and unprovoked seizures. Incidence 3%. Rare large CNVs were associated with epilepsy (array-based studies). The Canadian Epilepsy Network (CENet) conducted whole-genome sequencing of epilepsy patients to identify genetic variants that predispose individual to epilepsy or drug response.

4

Detecting CNV in Whole-Genome Sequencing

Read coverage variation

4

Detecting CNV in Whole-Genome Sequencing

Read coverage variation PopSV: Population-based approach

Use a set of reference experiments to detect abnormal patterns.

genomic window number of reads mapped

sample reference tested

5

PopSV’s workflow

6

PopSV is more sensitive than other methods

Twin dataset, normal/tumor cancer dataset and RT-PCR validation.

• LUMPY

CNVnator cn.MOPS FREEC PopSV 100 200 300

number of replicated calls per sample

• LUMPY

CNVnator cn.MOPS FREEC PopSV 0.00 0.25 0.50 0.75 1.00

proportion of replicated calls per sample

7

Application to the CENet dataset

7

Application to the CENet dataset

Frequency from 5 public WGS-derived SV databases. Rare means frequency < 1% in all 5 databases.

2000 4000 6000 8000 0.01 0.1 0.25 0.5 1

maximum frequency in the public databases CNV

rare common

8

Slight enrichment of rare CNVs in exons

all CNVs rare CNVs

• **
• **

**

0.6 0.8 1.0 all genes LoF intolerant genes all genes LoF intolerant genes

fold−enrichment

controls patients

fold-enrichment: how many CNVs overlap an exon compared to expected by chance. Loss-of-Function intolerant genes from ExAC consortium.

9

Rare exonic CNVs are more recurrent in the epilepsy cohort

• 0.000

0.025 0.050 0.075 2+ 3+ 4+ 5+ 6+ 7+ 8+ 9+ 10+

CNV recurrence proportion of rare exonic CNVs

• controls

patients

10

Putatively pathogenic exonic CNVs

10

Putatively pathogenic exonic CNVs

8/21 affect a known epilepsy-associated gene (Ran NAR 2015).

93.3 mb 93.4 mb 93.5 mb

Gene

CHD2

CNV

CNET0119 CNET0130 CNET0143 1000 1500 2000

Coverage

• CNET0119

CNET0130 CNET0143

Two recurrent CNVs were replicated in an additional cohort (325 patients and 380 controls).

11

Rare non-coding CNVs enriched close to epilepsy-associated genes

50 100 150 100 200 300

distance to nearest epilepsy exon (kb) cumulative affected samples

controls patients

12

Even more if in enhancers of the epilepsy gene

20 40 60 80 100 200 300

distance to nearest epilepsy exon (kb) cumulative affected samples

controls patients

Enhancer: eQTL or DNase site associated with the epilepsy gene.

13

Conclusions

Rare exonic CNVs are enriched and more recurrent in epilepsy patients compared to controls. Identified putatively pathogenic exonic CNVs, some replicated in an additional cohort. Rare non-coding CNVs are enriched close to epilepsy-associated genes. We show the importance of small and non-coding CNVs in epilepsy. Comprehensive profiling of CNVs could help explain a larger fraction of epilepsy cases.

14

Acknowledgment

Poster 8

Guillaume Bourque Pascale Marquis Mathieu Bourgey Louis Letourneau Francois Lefebvre Eric Audemard Toby Hocking Patricia Goerner-Potvin Simon Gravel Mathieu Blanchette Simon L. Girard Guy Rouleau Dan Spiegelman Alexandre Dionne-Laporte Jacques L. Michaud Fadi Hamdan Patrick Cossette Caroline Meloche Maxime Cadieux-Dion Micheline Gravel Ron G. Lafreniere Michel Boivin Danielle M. Andrade Cyrus Boelman Berge A. Minassian

15

16

Technical bias in WGS

WGS simulated shuffled 1200 1300 1400 1500 1600 1700

bin inter−sample mean coverage

WGS simulated shuffled 100 200

bin inter−sample standard deviation

17

Technical bias in WGS

• −0.10

−0.05 0.00 0.05

sample propotion of the studied genome coverage

• highest

lowest

18

Twin pedigree concordance

• 0.00

0.25 0.50 0.75 1.00 10 20 30 40

number of groups derived from CNV clustering Rand index using pedigree information method

• LUMPY

CNVnator cn.MOPS FREEC PopSV

clustering linkage

• average

complete Ward

19

RT-PCR validation rates

Region Validation rate Total 151 0.907 CNV type Deletion 102 0.902 Duplication 49 0.918 Frequency in database 26 0.923 (0, 0.01] 24 0.833 (0.01, 1] 101 0.921 Size (Kbp) < 20 73 0.849 (20, 100] 38 0.974 > 100 40 0.950

20

Size distribution

200 400 600 10 20 30 WGS array <10 10−50 50−100 100−200 200−500 >500

CNV size (Kbp) average number of CNV per sample

21

Large CNV enrichment in epilepsy patients

• all CNVs

rare CNVs all genes LoF intolerant genes all genes LoF intolerant genes 0.4 0.6 0.8 1.0

fold−enrichment

controls patients

22

Non-coding CNVs of high interest

23

Rare deletions enriched in epilepsy-associated genes

• 0.0

0.5 1.0 1.5 0.0 0.5 1.0 1.5 deletion duplication 1e−04 0.001 0.01 0.1 1

CNV frequency in public databases fold−enrichment

• controls

patients twins

P−value

• <0.05

>0.05

24

Rare deletions enriched in epilepsy-associated genes

• bserved

500 1000 1500 5 10 15 20

number of epilepsy genes among sampled genes number of sampling sampling

all genes size−controlled genes

Genes hit deletions never seen in public databases