Moderní metody analýzy genomu
Bioinformatika II
Karol Pál
Brno, 25.11.2016
Recap	from	last	lecture
• NGS	data	analysis
• Commercial	software	vs	“In	house”	pipeline	
• (Different	license	for	different	kind	of	experiment)	
HGMD
Reference Variants
Pre	alignment	
Quality	Control
dbSNP
Alignment	
Mapping
Variant	Calling
Annotation
Post	alignment	
processing+
IGV	(Showing	a	bam	file)
Experiment	design
• Modifications	to	the	basic	pipeline depending	on	input	material	and	
expected	output
• Next-gen Sequencing
• Whole Genome Sequencing (WGS)
• Targeted	Sequencing
• Whole	Exome Sequencing
• Gene	panels
• PCR	based
• RNA	Sequencing
• 3rd	generation	Sequencing
• Various	applications
Exome Sequencing	(WES)
• Input	material	:	
• Coding	regions	+	small	RNA’s
• Represents	1%	of	DNA	(human)
• Coverage	~	80x
• Scenarios:
• Single	individual
• Family	with	phenotype
• Paired	cancer	+	healthy	tissue	from	one	individual
CLINVAR
HGMD
WES	single	individual
EXAC
Variants
dbSNP
Annotation
Expected	output:
• (rare)	germline variants	
• ExAC
• ESP6500
• Kaviar
• dbSNP
• Inherited	disease
• HGMD
• CLINVAR
WES	Family	with	phenotype
Gemini
Expected	output:	find	disease	causing	mutations
GEMINI	=	SQLite	database
• SQL	queries
• Preformatted	queries	for	Autosomal	Dominant	
and	Autosomal	Recessive	phenotypes
WES	paired	samples	(healthy	+	cancer)
COSMIC
Variants
Variant	Calling
Annototion
Expected	output: Somatic	mutations
WES	not	ideal	(low	sensitivity)
Mutect:	
• Variant	caller	specialized	for	somatic	point	
mutations	in	cancer	genomes
• Takes	two	bam	files	as	input
Targeted	sequencing	PCR	+	Panels
• Higher	coverage	(up	to	10	000x)
• High	sensitivity	- somatic	variants	VAF	up	to	0.1%
• Methods	used	in	diagnostics
• CZECANCA	(panel)	targeting	219	cancer	susceptibility	genes
• BRCA	(PCR)
Targeted	sequencing	PCR	based	
Targeted	Sequencing
Reference
Post	alignment	
processing
Pre	alignment	
Quality	Control
• Trim	primer	sequences
• Do	not	remove	PCR	duplicates	
(PCR	based)
RNA	Seq
• Input	material:	cDNA
• Poly	A	tailing	
• Specific	capture
• Possible	outputs:	
• Expression	levels	(RPKM,FPKM	 or	tag	counting)
• Structural	variants
• (Variant	calling)	
RNA	Seq gapped	alignment
Reference
Variant	Calling
Gapped	alignment	vs.	alignment	to	“transcriptome”
RNA	Seq gapped	alignment
WGS	structural	variants
Methods	used
• Read	counts	(Coverage)	can	be	used	for	WES	
• Read	pair	(span	and	orientation	of	reads)
• Split	reads	(breakpoints)
• Assembly
Nanopore
• Long	reads	high	error	rate
• Scaffold	for	de	novo	assembly
• Transcripts
• Presence	of	pathogens
Thank	you	for	your	attention
links
• http://exac.broadinstitute.org/
• http://evs.gs.washington.edu/EVS/
• http://db.systemsbiology.net/kaviar/
• https://gemini.readthedocs.io/en/latest/
• http://archive.broadinstitute.org/cancer/cga/mutect
• https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4479793/