You discovered a novel bacterium of which the sequence is completely unknown. What NGS technology
would you prefer if you want to know the complete sequence of this novel organism.
Long read sequencing (for whole genome) and not make it too expensive: nanopore or bionano for whole
genome mapping. Looking at ion current: illumina; short reads (especially if repetitive sequences, however less
in bacteria compared to humans). With nanopore you could do long reads (has the longest reads) so nanopore
sequencing is the best option (longest reads). Nanopore because of the throughput, accuracy on PacBio is
better.
He would suggests: you have to write an answer on exam and argument why you choose this!!
Nanopore: because its cheaper and the longest reads (less acurate, need to sequence deep enough)
PacBio: good accuracy
Illumina: short reads no, because you don’t know the sequence yet so for variants you could get
into trouble. You don’t know if you have 1 or 2 repeats. (long read illumina could be possible; since
there is no sequence to align with).
Bionano: it’s not a sequencing technique but a mapping technique.
At exam: we assume ideal situation so unlimited amount of money.
You are working for a biotech company that commercializes proteins as well as metabolites for different
applications, mainly industrial. They are interested in a certain bacterium that seems to have a unique
metabolic characteristic. From earlier work, they suspect that the bacterium has a certain enzyme that
catalyzes a reaction that converts a certain cheap metabolite (that the bacterium takes up from the
environment) into a valuable end product. They suspect this because they were able to identify several
mutation strains that are not able to do the conversion. The company has an easy and highly automated
assay to detect the valuable end product. They have hopes to develop a business model to sell this enzyme
in the future. They have a number of questions and ask your advice.
How can they prove their hypothesis that the reaction converting the cheap metabolite into the valuable one is
a one enzyme reaction?
Different mutant strains: do a whole genome sequencing of all the mutants and the wild types to make sure
there are no other variants (and no stop mutations in enzymes) so to be sure that it’s just one enzyme causing
this. (or look for mutants in RNA sequencing, extract RNA converts in cDNA by NGS or …) In a higher organism
would be more difficult, nonsense mediated decay could destroy. One enzyme <-> different RNAs.
Determine if it’s a one or multiple if we find a mutation in the same gene of all the mutant strains (highly
likely that it’s the same enzyme) always an inactivating mutation in the same gene (this way we get the
whole sequence)
Would you sequence just one mutant? Or would you sequence more?
More would be necessary to validate. You need to find it in more mutants to be sure that it’s the same
enzymes in all the mutants.
How could they produce the enzyme in a cost-efficient manner?
Cloning in an expression vector and bring the expression vector into the bacteria. Bacterial enzymes can be
made via E. coli.
, What bacterium would you recommend?
E. coli because it’s a simple solution/bacterium. Give proper argumentations always!!!!
Imagine that for the treatment of a disease, patients are injected with a hormone. However, it turns out that
about 10% of the patients develop severe adverse side effects after the treatment. How could you evaluate
whether the gene encoding the receptor for that hormone could explain the different in effects observed in
patients?
First look at hormone, what could we do to explain these side effects?
Look at polymorphisms at the receptor; you can sequence the receptor. Polymorphisms in the receptor
between which groups? Between polymorphisms present in the ones with side effects and polymorphisms
ones without side effects. (expression of the receptor, in the patients with and without side effects in blood
samples).
Transcriptomics and compare patients with a control group, however we need the correct tissue, need to know
where it is expressed. No ethical clearance, if we test in mice they have side effects. Make IPSCs for the tissue
needed to sample. (take fibroblasts, take a small skin sample can make IPSCs).
It doesn’t have to be the hormone receptor, could also be something in the metabolism for example.
Or just normal genome sequencing (compare sequences between the two groups) look at sequence and
coding mutations.
Suppose we don’t find them, what do we do?
Association studies on large populations. In genomics, a genome-wide association study (GWA study,
or GWAS), is an observational study of a genome-wide set of genetic variants in different individuals to see if
any variant is associated with a trait. GWA studies typically focus on associations between single-nucleotide
polymorphisms (SNPs) (coding mutations) and traits like major human diseases, but can equally be applied to
any other genetic variants and any other organisms. (however risk of an adverse side effect; more difficult in
reality).
You are doing a PCR in the lab, and you try to amplify a part of a collagen gene, as a part of your diagnostics
for a new point mutation in a connective tissue. You ordered primers and a standard PCR. However, you are
getting several bands. How would you proceed in order to make it work? Start easy.
Primers are a little bit to aspecific so annealing goes wrong: look for more specific primers.
Gradient temperature to see what temperature is optimal. (higher annealing temperature)
(Check primer for secondary structures (hairpins or something), then probably some bands are from
that new primers)
Hot-start PCR cause that might work. (a method which prevents DNA polymerase extension at lower
temperature to prevent non-specific binding to minimize yield loss; enzyme only becomes active after
heating).
Nested primers, because you perform PCR twice with 2 different pairs of primers.
o Take first PCR product dilute away the original primers add new primers and do the
second PCR (be carefull for contamination)
would you prefer if you want to know the complete sequence of this novel organism.
Long read sequencing (for whole genome) and not make it too expensive: nanopore or bionano for whole
genome mapping. Looking at ion current: illumina; short reads (especially if repetitive sequences, however less
in bacteria compared to humans). With nanopore you could do long reads (has the longest reads) so nanopore
sequencing is the best option (longest reads). Nanopore because of the throughput, accuracy on PacBio is
better.
He would suggests: you have to write an answer on exam and argument why you choose this!!
Nanopore: because its cheaper and the longest reads (less acurate, need to sequence deep enough)
PacBio: good accuracy
Illumina: short reads no, because you don’t know the sequence yet so for variants you could get
into trouble. You don’t know if you have 1 or 2 repeats. (long read illumina could be possible; since
there is no sequence to align with).
Bionano: it’s not a sequencing technique but a mapping technique.
At exam: we assume ideal situation so unlimited amount of money.
You are working for a biotech company that commercializes proteins as well as metabolites for different
applications, mainly industrial. They are interested in a certain bacterium that seems to have a unique
metabolic characteristic. From earlier work, they suspect that the bacterium has a certain enzyme that
catalyzes a reaction that converts a certain cheap metabolite (that the bacterium takes up from the
environment) into a valuable end product. They suspect this because they were able to identify several
mutation strains that are not able to do the conversion. The company has an easy and highly automated
assay to detect the valuable end product. They have hopes to develop a business model to sell this enzyme
in the future. They have a number of questions and ask your advice.
How can they prove their hypothesis that the reaction converting the cheap metabolite into the valuable one is
a one enzyme reaction?
Different mutant strains: do a whole genome sequencing of all the mutants and the wild types to make sure
there are no other variants (and no stop mutations in enzymes) so to be sure that it’s just one enzyme causing
this. (or look for mutants in RNA sequencing, extract RNA converts in cDNA by NGS or …) In a higher organism
would be more difficult, nonsense mediated decay could destroy. One enzyme <-> different RNAs.
Determine if it’s a one or multiple if we find a mutation in the same gene of all the mutant strains (highly
likely that it’s the same enzyme) always an inactivating mutation in the same gene (this way we get the
whole sequence)
Would you sequence just one mutant? Or would you sequence more?
More would be necessary to validate. You need to find it in more mutants to be sure that it’s the same
enzymes in all the mutants.
How could they produce the enzyme in a cost-efficient manner?
Cloning in an expression vector and bring the expression vector into the bacteria. Bacterial enzymes can be
made via E. coli.
, What bacterium would you recommend?
E. coli because it’s a simple solution/bacterium. Give proper argumentations always!!!!
Imagine that for the treatment of a disease, patients are injected with a hormone. However, it turns out that
about 10% of the patients develop severe adverse side effects after the treatment. How could you evaluate
whether the gene encoding the receptor for that hormone could explain the different in effects observed in
patients?
First look at hormone, what could we do to explain these side effects?
Look at polymorphisms at the receptor; you can sequence the receptor. Polymorphisms in the receptor
between which groups? Between polymorphisms present in the ones with side effects and polymorphisms
ones without side effects. (expression of the receptor, in the patients with and without side effects in blood
samples).
Transcriptomics and compare patients with a control group, however we need the correct tissue, need to know
where it is expressed. No ethical clearance, if we test in mice they have side effects. Make IPSCs for the tissue
needed to sample. (take fibroblasts, take a small skin sample can make IPSCs).
It doesn’t have to be the hormone receptor, could also be something in the metabolism for example.
Or just normal genome sequencing (compare sequences between the two groups) look at sequence and
coding mutations.
Suppose we don’t find them, what do we do?
Association studies on large populations. In genomics, a genome-wide association study (GWA study,
or GWAS), is an observational study of a genome-wide set of genetic variants in different individuals to see if
any variant is associated with a trait. GWA studies typically focus on associations between single-nucleotide
polymorphisms (SNPs) (coding mutations) and traits like major human diseases, but can equally be applied to
any other genetic variants and any other organisms. (however risk of an adverse side effect; more difficult in
reality).
You are doing a PCR in the lab, and you try to amplify a part of a collagen gene, as a part of your diagnostics
for a new point mutation in a connective tissue. You ordered primers and a standard PCR. However, you are
getting several bands. How would you proceed in order to make it work? Start easy.
Primers are a little bit to aspecific so annealing goes wrong: look for more specific primers.
Gradient temperature to see what temperature is optimal. (higher annealing temperature)
(Check primer for secondary structures (hairpins or something), then probably some bands are from
that new primers)
Hot-start PCR cause that might work. (a method which prevents DNA polymerase extension at lower
temperature to prevent non-specific binding to minimize yield loss; enzyme only becomes active after
heating).
Nested primers, because you perform PCR twice with 2 different pairs of primers.
o Take first PCR product dilute away the original primers add new primers and do the
second PCR (be carefull for contamination)
