17.05.26
Lecture 3 - Genomics
* What is Genomics?
► We should remember two things.
- 'T' = 7 billion persons (species diversity) + 7 billion bases (variability in one person)
► Every life or species have genome, and variation of each genome is different.
* GEP/T graph.
► Geno + Enviro -> Pheno/Trait (almost cases are disease)
- Prediction is conducted by experiment, bioinformatics, etc.
► The challenge : As life is so complicated, prediction is so hard. We have to consider environmetal complexity with genomic complexity.
- Possible solution : using AI based on big data.
► Solutions
(1) Exact large amount of data.
- It is so important.
- typing (cheap and efficient) : simple test, approximation (not 100% prediction).
- sequencing (ultimate) : almost 100% prediction, digital device, it can have error (we don't know reason of error), ex) hardware.
(2) Need principle, law, and algorithms.
- We need to know related principle when we face to some facts.
(3) Big computer.
* Genome sequencing.
► 2nd generation : multiple molecule -> approximation
► 3rd generation : single molecule
► 4th generation : nanopore + membrane
► Base calling : computer notifies types of bases.
► NSG Data analysis : mapping based on reference gene.
* Close species comparative genomics.
► If we sequencing genome of new species, we have to find and get right right samples.
- we should pick sample which has interesting phenotype.
- And then, set up experimental design.
► Gene assembly : assemble gene fragments after putting the known gene between the intereted gene fragments.
- expensive
- After assmbling genome of one species, we can use it as reference genome.
► MRCA : most recent common ancestor -> we can predict
