Dalton - Genomic Testing in the NHS Flashcards
What pathway do samples take in the NHS from being taken to a report being made?
- assessment of test (input of clinical and/or diagnostic info)
- extraction of DNA/RNA/chromosome preps (or all)
- then either direct mutation analysis, genetic mutation detection or linkage analysis
- get results
- report written (so useful to patient)
What does the clinical pathway of a sample start and end with?
- the patient
When would direct mutation analysis be carried out on a sample undergoing genomic testing?
- if know what mutation is, eg. if in family already
WHen would genetic mutation detection be carried out on a sample undergoing genomic testing?
- if query a certain condition but don’t know specific mutation
- eg. if query CF then check CFTR gene
- have to have some understanding of clinical presentation to choose right test
What is also important to consider when deciding what tests to do on a sample?
- what patient has consented to
What is the referral reason important for?
- to decide what to do w/ sample
- eg. may be query CF, or may be more general
What types of sample can be taken prenatally?
- pre-implantation genetic diagnosis (PGD)
- free fetal DNA
- chronic villus sampling
- amniocentesis
- placental biopsy
- fetal blood sampling
What types of samples can be taken postnatally (and through to adulthood)
- cord blood
- dried blood spot
- salivary brush
- bone marrow
- cell free (tumour) DNA
- skin biopsy
- venipuncture
- post mortem
What is PGD and what can it be used for currently?
- take 1 cell from 8
- currently can only test specific region so have to know what looking for
What is the future for PGD?
- will be able to test whole genome eventually
- but should we?
What is free fetal DNA, and how is it sampled?
- DNA that circulates freely in maternal DNA
- sampled by venipuncture
When is chronic villus sampling taken, and what can happen if its done too early?
- 11 weeks
- clubbing
What does an amniocentesis sample?
- sampling amniotic fluid surrounding fetus
When can placental biopsies be taken?
- any time after 15 weeks
When do RBCs contain DNA?
- is fetuses and newborns up to 6 months
What sample is taken from all newborns in England?
- dried blood spot
What is the issue w/ amniocentesis?
- risk of miscarriage, but v low risk
- having termination at this time is induced labour
How are genes labelled, ie. nts, exons, etc.?
DIAG
Do cDNA and gDNA contain introns?
- gDNA does
- cDNA is cop of RNA so does not
What kind of brackets are used in mutation nomenclature to indicate predicted protein change?
- ( )
In mutation nomenclature how would a point change be written for cDNA, eg. C changed to T at position 145?
- c. 145 C>T
In mutation nomenclature how would predicted protein be written, for eg. Arg to stop at position 49, or to Leu?
- p. (Arg49*)
- p. (Arg49Leu)
What is a nonsense mutation, and how likely is this to be pathogenic?
- AA to stop
- more likely to be pathogenic (but not always), as disrupt prod of protein though nonsense mediated decay
What is a missense mutation, and how likely is this to be pathogenic?
- changes AA
- much harder to predict, could be pathogenic or not
In mutation nomenclature how would a change in the intron be written for cDNA, eg. for G changed to T at position in intron before 51 in the exon?
- don’t want to inc gDNA numbering as well, so use last no, from cDNA, ie -1 means go back 1 towards 5’ end (can do from either end)
- eg. c. 51-1G>T
In mutation nomenclature how would you write an unknown predicted protein change, and why would we not know?
- p. (?)
- if splice site mutation don’t know what it does, as how cell deals w/ it can vary –> exon may be removed or intron inc
In mutation nomenclature how would a deletion be written for cDNA, eg. a TG deletion at position 104 to 105?
- c. 104_105delTG
What is the effect of a 2 base deletion in mammals?
- disrupts ORF as not multiple of 3
- generally get stop w/in 100 AAs, but usually w/in 10 and often immediately
In mutation nomenclature how would the predicted protein change be written for a 2 base deletion, if eg. Cys changed to Tyr at position 34, and frameshift of 12 AAs?
- p. (Cys34Tyrfs*12)
- frameshift is at 12AAs from and inc 1st AA changed –> get stop
Do frameshift mutations always mean all AAs afterwards are changed?
- no, sometimes can get some AAs after where maintained
In mutation nomenclature how would an insertion be written for cDNA, eg. of A at position 104?
- c. 104insA
In mutation nomenclature how would how would a duplication be written for cDNA, eg. of T at position 104?
- c. 104dupT
In mutation nomenclature how are larger deletions written for cDNA, eg. of bases 94-97?
- c. 94_97del4
In mutation nomenclature how would deletion of an exon be written for cDNA, if don’t know exactly what’s been lost as unsure where breakpoint is, eg. deletion of part of exon 2, between positions 51 and 141?
- c. 51-?_141+?(del exon 2)
In mutation nomenclature how would a 3 base deletion be written for cDNA, eg. of TTT at 133 to 135?
- c. 133_135delTTT
In mutation nomenclature how would a 3 base deletion be written for predicted protein change, eg. was Phe at position 45?
- p. (Phe45del)
What is generally the effect of 3 base deletions?
- not pathogenic and quite common, as doesn’t alt ORF
What is the nomenclature for a known genotype?
- [genotype]
What is the nomenclature for 2 diff alleles in trans (give eg.)?
- p. [(Phe45Leu)] ; [(Arg49*)]
- for AAs write name then no. but for nts write no. then name
What is the nomenclature for same allele in cis (give eg.)?
- p. [(Phe45Leu) ; (Arg49*)]
- remove internal [ ] as know on same allele
What is the nomenclature if unsure whether on cis or trans?
- p. [(Phe45Leu) (;) (Arg47*)]
- ; in brackets as don’t know
What is the nomenclature for heterozygous changes in a limited screen?
- c. [94_97del4] ; [94_97 =]
- on other allele only looked at these bases to see if WT, can’t say rest is WT as we don’t know
What is the nomenclature for heterozygous changes in a full screen?
- c. [94_97del4] ; [=]
- no no. in front of = as looked at whole gene on other allele and found to be WT
What is the nomenclature for hemizygous changes?
- c. [135C>A] ; [0]
What is the nomenclature for somatic changes (mixed cell pop)?
- c. [51-1G>T / =]
- / means allele diff in diff cells
How can you define mutation (if germline and if somatic)?
- if in germline = disease causing alteration that can be inherited
- if somatic = disease causing alteration that has occurred after meiosis
What is a polymorphism?
- alteration which is w/o effect or advantageous, and can be inherited
Does frequency of occurrence define whether something is a mutation or polymorphism, what can be looked at instead?
- no, polymorphisms can be rare and mutations can be common
- incidence = look at disease freq, ie. if have common change and rare disease then assoc not likely, so prob polymorphism
When are nonsense mutations less likely to be pathogenic, and what happens?
- when near end of ORF
- as need big gap between incorrect and normal stop codon for it to be targeted by NMD
- so get abnormal peptide, which may or may not be pathogenic and cause phenotypic effect
Why do we get splicing variants, and what is the result of this?
- can result in exon being deleted or intron inc
- if intron inc can result in stop and NMD, due to alt ORF
- if exon del can be in frame, so n stop –> get milder phenotype or none
- some genes spliced diff in diff places –> so not only have to know about what splice variant is and does, but also good knowledge of gene itself and how expressed in diff tissues
What is an eg. of genes spliced diff in diff place, and how is this signif when looking for variant causing symptoms?
- glycogen storage genes in muscle and liver
- so if symptoms muscular, but variant positioned in exon only transcribed and translated in liver, then prob not mutation
How can you tell if something is a mutation or a polymorphism - ie. what can we look at?
- look at freq vs incidence
- type of mutation
- splicing variants
- look at degree of conservation
- functional info
- structural info
- functional studies in vitro
- functional studies in vivo (not done)
- clinical info
How can we look at degree of conservation to see if a variant is a mutation of polymorphism?
- look at how well particular base/AA/nt conserved
- is it conserved all the way through evo history?
- does it vary when move from humans to close relatives (eg. chimps)?
- use multiple seq alignment (MSA)
What functional info can be looked at to see if a variant is a mutation of polymorphism?
- motifs (eg. does it bind ATP?)
- functional domains
How can functional studies be carried out in vitro to see if a variant is a mutation of polymorphism?
- eg. might be enz and can look at activity or do mutagenesis
- or use Ts mutants
What clinical info can be looked at to see if a variant is a mutation of polymorphism?
- go back to info on patient
- does change we’ve seen in gene assoc w/ symptoms in patient (if not then not helpful to patient and prob not relevant)
What types of software tools and packages are there for looking at variants (and often meta analyses)?
- machine learning methods
- protein seq and structure based methods (algorithms that looks at predicted/known protein structure)
- seq and evolutionary conservation based methods
- analysis and collation packages
How are variants classified?
- benign -1
- likely benign -2
- unknown signif -3
- likely pathogenic -4
- pathogenic -5
What does a variant classification of -1 mean?
- v certain is polymorphism
- high pop freq
- not linked to patients phenotype
What does a variant classification of -2 mean?
- some freq data
- in silico predicts neutral
- not in functional domain or likely to impact structure
- few/no pathogenic mutations in region
- but can’t be certain not pathogenic
What does a variant classification of -3 mean?
- conflicting predictions from in silico tools
- no freq data or lit
What does a variant classification of -4 mean?
- no/low pop freq
- suggestion that mutations in region identified before
- pathogenic in silico prediction
- disease specific lit
- but can’t be certain pathogenic
What does a variant classification of -5 mean?
- segregates w/ disease
- functional study
- all evidence in agreement
- def pathogenic
What classification are lots of variants found, and why is this a problem?
- unknown signif (-3), esp in genes don’t know lot about
- as can’t give patient info either way
Who’s guidelines were adopted for variant interpretation?
- American College of Medical Genetics (ACGS)
What does linkage analysis involve following the inheritance of?
- following inheritance of particular parts of chromosome through families
What does linkage use as markers?
- natural polymorphisms w/in DNA
What samples do we still use linkage for (instead of direct mutation analysis)?
- PGD
When is linkage esp useful, and how?
- in patients who don’t want to know things
- can track variants and how segregate to see who’s inherited what chromosome from who
- look at microsatellites (/variable nt triplet repeats)
How many variants do people have in their DNA?
- approx 5 mil
What about microsatellites varies?
- no. present in indiv
What are looking at microsatellites used in?
- DNA fingerprinting
What type of microsatellites are most desirable for analysis and why?
- intragenic
- problem if not due to recomb
- greater distance from gene = greater chance of recomb, but relationship not linear (due to recomb hotspots)
Why are SNPs less informative than microsatellites?
- 1 of 2 bps
- can be intronic or exonic
What are the methodologies of PCR based technologies?
- amp of target DNA/cDNA
(input of sequencing –> pyroseq/Sanger/NGS) - electrophoresis
- visualisation (how is dep on size)
What happens to PCR if part where primers bind del?
- won’t get amp
How can changes in DNA be identified by PCR?
- put primers where deletion suspected, and if get amp then DNA is there
What is the fundamental problems w/ this method of using PCR to identify changes in DNA?
- if looking for del on X chromosome and looking at male and PCR doesn’t work –> is this because they’re male?
- and if look at female w/ 1 normal X chromosome?
- so have to be able to distinguish between half amount of PCR product and normal amount
- but usually take PCR to completion so no longer prop to to starting material
- therefore important to consider dosage
What are other poss problems w/ using PCR to identify changes in DNA?
- doesn’t always work (need good control)
- contam (esp from PCR products)
- pol can introd mutations
- non specific binding if primers not specific enough
- repeated seqs in gDNA and pseudogenes etc.
What are the advs of direct mutation tests, eg. ARMS (amp refractory mutation system)?
- cheap
- quick
- don’t get unwanted info
- can do lots in multiplex
How is ARMS carried out?
- design primer complementary to WT (control) and 1 to mutant
- so can design primer and get PCR products formed only when mutation present
- put stacker on end of 1 primer to make them diff size so can distinguish between products (as will be 2 bands)
- can test for multiple mutations at once but have to know seq info and what looking for
- have to remove pol proofreading activity as will correct extra base
What is the important thing to remember about ARMS?
- targeted to particular mutation
Why doesn’t the NHS just use genomic screen for everything, as will def find what looking for?
- issues of finding out things didn’t want to know
- cost (and responsibility to deliver value for money service as publicly funded)
Why is automated Sanger seq used, even though so old?
- v robust –> often seen as gold standard as well worked and well understood
How does Sanger seq work, and how does it provide the seq?
- seq reaction w/ labelled and unlabelled dNTPS
- create ss template (invariably from PCR) by splitting PCR
- use seq primers to seq across, in presence of fluorescently labelled dNTPs (system able to detect these)
- when dNTP incorp into seq it terminates, so get group of products stopping at primer +1, primer +2 etc., w/ diff dNTP labels at that point
- carefully titrate, so get series of products that can run out through capillary gel, sep by size
- for every 1bp change know what end dNTP is as can detect fluorescence
What is the result of Sanger seq, ie. the trace, and how is it read?
- get trace differential, which takes 1st trace away from 2nd so can see what diff is in peak height and area
- no diff gives straight line, diff gives “bubble”
- can still get variations, as long as consistent w/in seq