Genomic DNA extraction
A silica membrane in a microcentrifuge tube: the lysate is mixed with guanidine and ethanol and spun through, the DNA adsorbs, two washes take the chaotrope away and a small volume of low-salt buffer elutes it. The default for genotyping and most molecular work.
Our ruleSimulations prepare you for the bench but don't replace hands-on practice.
Spin columns
A silica membrane in a microcentrifuge tube: the lysate is mixed with guanidine and ethanol and spun through, the DNA adsorbs, two washes take the chaotrope away and a small volume of low-salt buffer elutes it. The default for genotyping and most molecular work.
How it differs from the standard method
- Fast, reproducible and forgiving, with yields of 20–40 µg before the membrane saturates.
- The membrane has a hard binding capacity, so more input past about 25 mg of tissue returns almost no extra DNA and makes the purity worse.
- Spinning the lysate through shears the DNA somewhat, so a column prep typically caps out around 50 kb.
1Three problems, not one
Getting genomic DNA out of a cell is three separable jobs, and almost every extraction failure is one of them going wrong. First you have to break the cell and the nuclear envelope open, so the DNA is in solution at all.
2Why chaotropic salt makes DNA stick to glass
DNA's phosphate backbone is surrounded by an ordered shell of water, and silica carries a hydrated, weakly negative surface. Left alone, two hydrated negative surfaces do not touch.
Detergent and proteinase K
SDS or a similar detergent opens membranes; proteinase K digests nucleases and the protein bound to the DNA. Incomplete digestion shows as visible tissue, low yield and a low A260/A280.
Silica, beads or an organic phase
Chaotropic salt and ethanol for silica; PEG and salt for beads; phenol-chloroform partitioning then alcohol precipitation for the organic route.
Wash, dry, elute
Two washes to get the chaotrope out, a dry step to get the ethanol out, then a low-salt buffer to rehydrate the DNA and lift it off.
3What carry-over does downstream
- Guanidine salts inhibit polymerases and ligases, and absorb hard at 230 nm, so carry-over shows up as an A260/A230 well under 1.8 long before it shows up in A260/A280.
- Ethanol inhibits enzymes at under 1% v/v and makes the sample float out of a gel well. It gets into the eluate when the dry step is skipped.
- Phenol absorbs across 230–270 nm with a peak at 270, so a prep with phenol in it shows a 270 nm shoulder and a depressed A260/A230 while A260/A280 can look unremarkable.
- The diagnostic for an inhibitor rather than too little template is that diluting the sample 1:10 or 1:20 makes the reaction work better, not worse.
- RNA binds silica and beads as well as DNA does, so without an RNase step the absorbance concentration can be twice the real DNA concentration, and A260/A280 barely moves, because RNA's own ratio is only about 2.05.
4Fragment size is a result too
A human chromosome is tens to hundreds of megabases long, and nothing you do at the bench keeps it intact. What you can control is how much further you break it.
- Bead-to-sample ratio
- The volume of PEG/salt bead mix per volume of sample. It sets both the binding capacity available and the size below which fragments stay in the supernatant: fewer beads means a higher cut-off, so a low ratio is a deliberate size selection as well as a saving.
- Plant tissue
- Leaves add two problems animal tissue does not have: cell-wall polysaccharides, which are viscous, co-purify with DNA and inhibit enzymes, and polyphenols, which oxidise and bind DNA irreversibly. CTAB with high salt keeps polysaccharides in solution while DNA precipitates; PVP traps polyphenols before they oxidise.
Key terms
Practise these as flashcards →A salt such as guanidinium thiocyanate that disrupts ordered water, letting DNA adsorb onto silica. Also a strong enzyme inhibitor, so it must be washed out.
A robust serine protease active in detergent and at 56 °C. It digests nucleases and the protein bound to DNA, and is itself inactivated later by the chaotrope or by heat.
The white layer between the aqueous and organic phases of a phenol extraction: denatured protein with phenol trapped in it. Carrying it over puts both into your sample.
DNA kept long enough for long-read sequencing, conventionally an N50 above about 30–50 kb. It is defined by handling, not by chemistry.
The fragment length above which half the mass of the DNA lies. A more useful description of a size distribution than a mean, because mass is dominated by the long molecules.
Cetyltrimethylammonium bromide. In high salt it stays in solution with DNA; drop the salt and the CTAB-DNA complex precipitates, leaving polysaccharides behind.
The purity ratio that sees guanidine, phenol, polysaccharide and residual salt. Around 2.0–2.2 for a clean prep; well under 1.8 means something an enzyme will notice.
A pipette tip with the end cut off or moulded wide, used to reduce the shear rate at the orifice when moving long DNA.
When to use it
- Genotyping a mouse colony, a cell line or a cohort, where you need reliable PCR template from many samples and fragment length does not matter.
- Building a whole-genome or exome library, where you need an accurate dsDNA mass and an eluate clean enough for enzymatic steps.
- Long-read sequencing or optical mapping, where the prep has to be chosen and handled for fragment length first and yield second.
- Archiving DNA from a tissue, blood or plant collection, where stability in storage matters more than how fast the prep is.
Compared with alternatives
The next step every time: it tells you how much DNA you actually have, how pure it is and how long the fragments are. Prefer a dsDNA dye over absorbance whenever RNA might be present.
Prefer it when the target is a plasmid: alkaline lysis exploits the plasmid being small and supercoiled, which a genomic prep cannot do.
Prefer it when the target is RNA. The same guanidine chemistry is used, but with an acidic phenol step that keeps DNA in the organic phase, and everything has to be RNase-free.
The cheapest integrity check: intact genomic DNA runs as one high band, degraded DNA as a smear down the lane.
The application that makes the extraction method a real decision, because read length cannot exceed the fragment length you hand it.
Sub-techniques
Choosing one here switches the whole page, including the protocol and the lab, to it.
A silica membrane in a microcentrifuge tube: the lysate is mixed with guanidine and ethanol and spun through, the DNA adsorbs, two washes take the chaotrope away and a small volume of low-salt buffer elutes it. The default for genotyping and most molecular work.
- Fast, reproducible and forgiving, with yields of 20–40 µg before the membrane saturates.
- The membrane has a hard binding capacity, so more input past about 25 mg of tissue returns almost no extra DNA and makes the purity worse.
- Spinning the lysate through shears the DNA somewhat, so a column prep typically caps out around 50 kb.