How DNA extraction methods differ

DNA extraction is the decision about what you are willing to lose. Every method lyses a specimen, then separates nucleic acid from protein, lipid and the small molecules that would poison the next enzyme. Columns, beads, alcohol precipitation and organic solvents do that separation differently, so they return different yields, different fragment lengths and different leftovers. This page is a way to choose among those classes. It is a research explainer, not a kit insert and not a clinical sample protocol.

The downstream assay picks the winner. A short PCR can tolerate sheared DNA and still fail on a trace of ethanol. A long-read library can forgive a lower yield and still fail if the fragments were broken in the lysis. RNA co-purified by accident is a separate problem, taken up in protecting RNA during extraction.

Sourcing starts from the sample preparation catalogue. Send the specimen and the assay with the quote request. A catalogue family name is not a promise about a single formulation.

What every method is trying to do

Lysis has to open the protein and membrane around the DNA. Detergent, chaotrope, alkali, enzyme, heat or grinding each open a different matrix. Plant tissue needs a way past cell walls and polyphenols. Bacteria with tough walls need an enzyme or a mechanical step that mammalian cells may not. A buffer that works on a cultured cell pellet can leave a tissue chunk looking extracted when most of the DNA is still inside.

Once the DNA is free, the method has to keep it and discard the rest. Silica chemistry binds DNA in high chaotropic salt and releases it in low salt. Magnetic beads use a related bind-and-wash idea, or a crowding-agent chemistry that can also select by size. Alcohol precipitation can preserve long molecules and can also carry polysaccharides. Organic extraction partitions proteins into a phenol phase and leaves nucleic acid in the aqueous phase only if the pH matches that protocol.

Each method has a leftover it is bad at removing. Knowing that leftover is more useful than memorising a brand name.

How the classes differ in practice

A silica spin column is fast and has a finite binding capacity. Too much lysate, and DNA passes through while you record a low yield. The wash that removes chaotrope usually contains ethanol, and ethanol left in the eluate inhibits polymerases and ligases. A dry spin helps. Over-drying the membrane can make elution poor in the other direction. Elute in the buffer the next assay accepts. Tris-EDTA protects stored DNA and can also chelate magnesium the next enzyme needs, so storage buffer and assay buffer are allowed to differ.

Bead methods share that wash logic. A magnet replaces the column frit, which suits automation when the magnet and the tube match. Silica-paramagnetic beads behave like a column. Carboxyl beads used with a crowding agent and salt bind in a size-biased way: more crowding agent pulls in shorter fragments. That bias helps when you are removing primer-dimer and hurts when you wanted those short pieces. Beads do not repair a bad lysis.

Precipitation methods, including salting-out followed by alcohol, remain rational when you want long genomic DNA and can accept a pellet that is slow to dissolve. The pellet can be invisible, or mostly salt. A careful wash and patience during resuspension matter more than a vortex that shears the DNA you meant to keep long. Residual salt and ethanol are the usual reasons the next enzyme is quiet.

Organic extraction with phenol and chloroform is effective and hazardous. The aqueous layer holds DNA only under the pH of a DNA protocol. Acidic phenol recipes used for RNA move DNA the other way. The interphase holds protein, and pipetting through it carries contamination. Only laboratories that already have the cabinet, the waste path and the training should treat this as a default.

Crude lysates, including hot alkaline lysates and chelating-resin preps, can feed a hardy PCR and still be the wrong tool for quantity, cloning or a scarce sequencing target. Inhibitors travel with the crude material. A dilution is often the real purification step.

A workflow with branch points

Start from the specimen, not from the kit that happens to be in the freezer. Record input mass or cell number, because "low yield" is meaningless without it. Lyse until the matrix is actually dispersed. A clear lysate from tissue that still has a core is not a complete lysis.

If the yield is low and the input was high, suspect overload or incomplete lysis before you suspect the elution buffer. Split the lysate or repeat the lysis on a matched piece and see which change moves the number. If the input was tiny, you may simply be under the practical range of that chemistry.

If absorbance looks high and PCR fails, suspect inhibition. Dilute the eluate tenfold and try again. A recovery after dilution means inhibitors or too much DNA, not an absent target. A blank extraction — water or buffer taken through every wash — belongs in the same PCR. A band from the blank is contamination acquired during the prep, and the sample bands are not interpretable until that is solved. The logic is the same as in PCR controls and contamination control.

If the DNA must stay long, stop vortexing genomic preps and judge length on a gel or a fragment trace. A 260/280 ratio near the familiar DNA window is a hint about protein, not a length measurement and not an inhibitor test. RNA in the same tube inflates the 260 reading.

Failure modes worth naming

Brown or viscous plant eluates often still contain polyphenols or polysaccharides that bind DNA and the next polymerase. Use a plant-aimed method or a cleanup chosen for that contaminant.

A smear of very short DNA on a gel means the specimen was already dead and degraded, or the lysis was violent, or a nuclease stayed active because the chaotrope or the heat step never fully stopped it. Preserving specimen condition before lysis beats any column. Tissue left warm on a bench has already decided the fragment length.

If a downstream enzyme dies after an organic prep, assume solvent carry-over until a cleanup says otherwise. Guanidinium left by a skipped wash does the same quiet damage and often shows up as a poor 260/230 ratio. That ratio is another hint, not a verdict. Plasmid alkaline lysis is a different procedure from genomic extraction. Name the molecule you want before you name the kit.

Safety, climate and the enquiry

Chaotropic salts are harmful. Phenol and chloroform are worse. Follow the hazard note on the bottle you opened, and follow institutional waste rules. Specimens from people or infected animals stay inside the biosafety decision your committee already made. The WHO laboratory biosafety manual is a public reference for that institutional conversation. This page does not approve a containment level.

Washes dry slowly in humid air, so a column that looks dry can still hold ethanol, and the next polymerase stalls. A centrifuge that stops mid-run because the power dropped has not finished the separation. Restart only if you know which fraction is which.

EVRINTH can take a sourcing question. State the specimen, the input range, the fragment length you need, the downstream assay, and any inhibitor you already know is in the matrix. The nucleic acid isolation enquiry reference is a place to frame that question. It is an independent method reference. Ask whether a quotation is possible. Do not read it as a statement that a preparation service is already running. The sample preparation pathway connects extraction to the assays that consume the eluate.