ANOVA in Excel for Beginners: Mean, SD and Post-Hoc Tests

Meeting someone new follows a script. How are you? What do you do? Where did you study? Nobody skips it, because those small questions build the ground for anything more interesting. Data, I have come to think, deserves exactly the same courtesy.

Before you throw a statistical test at your results, you greet them. What is your average? How scattered are you? What do you look like on a graph? Only then do you ask the big question: are you really different from each other? This post walks through that conversation in Excel, using one control and three experimental conditions.

The walkthrough is adapted from the Adwoa Biotech YouTube tutorial. I have tightened a few statistical points along the way.

Step 1: Meet the average

Say you have seven readings for each of four groups: a control and three conditions. The mean gives every group one representative number. You add up all the values, divide by how many there are, and in Excel that is simply =AVERAGE(...).

One warning, though. When your values are wildly scattered, the mean becomes a poor spokesperson. For skewed data, the median (reported with the interquartile range) tells a more honest story.

Step 2: Ask about the spread

An average on its own is a sentence without context. Two groups can share the same mean and behave completely differently, so you also ask how far the readings wander from it. That is the standard deviation, and for a sample you want =STDEV.S(...).

You will often see the standard error of the mean (SEM) too, which is the SD divided by the square root of n. They answer different questions, so it helps to keep them apart:

• SD describes how scattered your data are.
• SEM describes how precisely you have pinned down the mean.

Whichever one you plot, say so in the figure legend. Your readers should never have to guess.

Step 3: Draw it

Our four groups are independent of one another, so a bar chart fits well. Had they been time points, I would reach for a line graph to show the progression instead.

In Excel, go to Insert and pick a bar chart. Then comes the part I enjoy most: decluttering. The rule is simple, and a little ruthless. If an element does not help communicate the data, it goes. Gridlines leave first, and the cleaner chart feels a bit like opening a window in a stuffy room.

After that, add what a stranger needs in order to read it:

• Axis titles that name the quantity and its units, for example "Absorbance (405 nm)", with "Experimental condition" on the x-axis.
• A title that tells the viewer what the chart is about.
• Error bars, with one catch. Excel's built-in "Standard Deviation" option calculates across the plotted bars rather than within each group, so the result is meaningless here. Choose Custom and point it at the SD (or SEM) cells you calculated yourself.

With small samples like n = 7, consider overlaying the individual data points as well. Bars can hide a surprising amount of drama.

Step 4: Are they really different?

For two groups, a t-test does the job, and in Excel the Welch version (unequal variance) is the safer default. With four groups, though, it is tempting to run a t-test on every pair. Resist that temptation.

Why? Four groups give six pairwise comparisons, and each test is another roll of the dice for a false positive. At α = 0.05 across six tests, the chance of at least one fluke is 1 − 0.95⁶, which is roughly 26%. That inflated error rate is exactly why we use analysis of variance (ANOVA): one test across all the groups, with the false-positive rate held at 5%.

Switching on the Analysis ToolPak

Excel keeps this tool behind a door, and nearly everyone looks for it in the wrong place at least once.

1. Go to File, then Options, then Add-ins.
2. At the bottom, set Manage to Excel Add-ins and click Go.
3. Tick Analysis ToolPak.

A Data Analysis button now appears on the Data tab. It lives under Add-ins in the options menu, not under Data, which is where the trap is.

Feeding it the right food

ANOVA does not want your means and SDs. It needs the raw values for every replicate, because the variance is precisely what the test is measuring. If your numbers come from a standard curve, convert each replicate to a concentration individually before you start.

• Put each group in its own column, with a heading.
• Paste values, not formulas, when moving data between cells.
• If you include headings, tick Labels in first row, otherwise Excel complains about non-numeric data.
• Choose an output range yourself, because Excel will not pick one for you.

Then select Anova: Single Factor. Alpha is set to 0.05 by default, which is the usual convention. A stricter threshold would be 0.01.

Reading the output

You get a summary table (count, sum, average and variance for each group) and an ANOVA table. The second one is the part that makes people nervous, so here is what each column means:

Term	What it means	In our example
SS	Sum of squares: total variability, split into between and within groups	Reported for each
df	Degrees of freedom. Between = groups − 1. Within = observations − groups	3 and 24 (28 − 4)
MS	Mean square: SS divided by df, an average amount of variation	Reported for each
F	MS between divided by MS within	One value
P-value	Probability of data this extreme if the null hypothesis were true	About 0.0019

Notice that there is only one p-value. That is because there is only one test. The "within groups" row is not a second result, it is simply the yardstick the F ratio is measured against.

What a p-value does and does not say

Our p of 0.0019 sits below 0.05, so we reject the null hypothesis that all the group means are equal. Two misreadings are very common, and both are worth avoiding:

• Alpha is not confidence in the null hypothesis. Setting α = 0.01 does not mean you are 99% sure nothing is happening. It means you accept a 1% false-positive rate when there truly is no effect.
• A p-value is not "the probability it was all chance." It is the probability of seeing data at least this extreme if the null hypothesis were true.

The catch: ANOVA does not say who

This is the part that gets skipped most often. ANOVA is an omnibus test. It tells you that at least one group differs, and then it shrugs. Did condition 1 beat the control? Are conditions 2 and 3 the odd pair? The test cannot say.

To find out, you need a post-hoc test:

• Tukey's HSD compares every pair of groups.
• Dunnett's test compares each condition against the control, which is often exactly what an experiment needs.

The ToolPak includes neither, so you will need other software such as R, GraphPad Prism or JASP.

Before you trust it

ANOVA assumes the observations are independent, the residuals are roughly normal, and the group variances are similar. With seven values per group, those assumptions are hard to verify. If they look shaky, the Kruskal-Wallis test is the non-parametric alternative, although Excel does not have it built in either.

The short version

1. Say hello: mean, SD (or median and IQR), and a clean chart.
2. Run ANOVA on the raw replicates, never on the summaries.
3. Read the p-value carefully.
4. Follow up with a post-hoc test to learn which groups differ.
5. Tell your readers whether the error bars show SD or SEM.

Video callout: Want to see the clicks in action? Watch the original walkthrough here: https://www.youtube.com/watch?v=3jYo1W6hyN0

Related Content

• Calculating log2 and log10 values

5% Haematocrit in P. falciparum Culture: Volumes and Why

When I was culturing parasites in the field, one number followed me everywhere: 5%. Five percent haematocrit. It sounds like a small, fussy detail, doesn’t it? Yet it quietly decides whether a culture thrives or collapses.

What haematocrit actually means

Haematocrit is simply the proportion of a suspension that is packed red blood cells. In a culture at 5% haematocrit, red cells make up a twentieth of the total volume, and everything else is culture medium. Think of it as a very thin cellular soup, mostly broth with a modest scattering of cells.

Why not more cells?

Plasmodium falciparum lives inside red blood cells, so you might assume that more cells means more parasites. In practice, a denser culture burns through nutrients quickly, acidifies the medium, and leaves you changing it far more often than you would like. In the field, where supplies are limited and the power can be unpredictable, that matters a great deal.

Why not fewer?

Too sparse, and merozoites struggle to find fresh cells to invade after the schizont bursts. Growth slows, and the parasitaemia you worked so hard to establish stalls. Five percent is a compromise: enough cells for efficient reinvasion, enough medium to keep conditions stable between changes.

What 5% looks like at the bench

Here are two culture sizes I used regularly. In both, the volume is the total culture volume, so the cells are part of it, not added on top.

• For a 5 mL culture: 250 µL of packed red cells (infected cells included) with 4.75 mL of complete medium.
• For a 10 mL culture: 500 µL of packed red cells (infected cells included) with 9.5 mL of complete medium.

If you work from a 50% haematocrit stock instead, double the cell volume: 500 µL of stock for a 5 mL culture, or 1 mL for a 10 mL culture. Top up with complete medium to the final volume.

A few habits that helped me

Always start from properly washed, packed cells, after a clean spin with the supernatant and buffy coat removed. Packed cells are viscous, so use a wide-bore or cut tip and pipette slowly. If you find small volumes of packed cells awkward to measure, make a more dilute stock first and work from that. It is easier to be accurate with a larger volume, and consistency from one passage to the next is what keeps a culture healthy.

The takeaway

Five percent is not magic, but it is a reliable starting point that balances parasite growth against nutrient supply. Once you understand why it works, you can adjust it sensibly when conditions call for something different.

Video callout: Want to see this in practice? Watch the walkthrough on the Adwoa Biotech YouTube channel: https://youtu.be/AF1bkbbgf9w?si=_9ICA7np-ECNXp4s

Related Content

• Hematocrit calculation for passaging
• RBC isolation protocol
• P. falciparum life cycle

MagMAX RNA Isolation Protocol for Tempus Blood RNA Tubes

 

MagMAX for Stabilized Blood Tubes RNA Isolation Kit Protocol for Tempus Tubes (Cat. 4451893)

This is the manufacturer's protocol for purifying total RNA, including microRNA, from human whole blood collected in Tempus Blood RNA Tubes, paraphrased from the Applied Biosystems protocol (Part No. 4452007, Rev. A). It works with magnetic beads instead of spin columns, and you can run it in 1.5 mL tubes or in the 96-well Processing Plate supplied with the kit. Expect roughly 3 to 25 µg of RNA from 3 mL of blood.

Before you start

• Draw 3 mL of blood into the Tempus tube and shake or vortex it for 10 seconds straight away. Poor mixing means poor stabilization and microclots.
• Stabilized blood keeps for up to 5 days at 18 to 25 °C, up to 7 days at 4 °C, and long term at −20 to −80 °C (keep it off direct contact with dry ice).
• Prepare Wash Solution 1 by adding 8 mL of 100% isopropanol to each bottle of concentrate. Prepare Wash Solution 2 by adding 24 mL of 100% ethanol to the concentrate. Invert 3 to 4 times, mark the bottles, and store at room temperature. Each isolation uses 300 µL of each.
• You need a refrigerated swing-bucket centrifuge with 50 mL tube adapters that can reach 5,000 × g at 4 °C, a magnetic stand, and either a vortexer with a 1.5 mL tube adapter or an orbital plate shaker.
• Vortex adapter settings used by the manufacturer (scale of 1 to 10): 4 to 5 for binding and washes, 6 to 7 for digestion and elution. Orbital plate shaker settings: 6 to 7 for binding and washes, 8 to 9 for digestion, drying and elution. Test your own equipment with water so nothing spills.
• Wear gloves, use RNase-free tips, and wipe the bench and pipettes with an RNase decontamination solution.

Dilute the stabilized blood

1. If the sample is frozen, thaw the Tempus tube on ice for 10 to 30 minutes. Keep samples on ice as much as possible.
2. Pour the whole contents into a labeled 50 mL conical tube.
3. Add 3 mL of Tempus 1× PBS to reach 12 mL. If you collected less than 3 mL of blood, top up with PBS to 12 mL, otherwise yields drop sharply.
4. Cap tightly and vortex at vigorous speed for 30 seconds. Froth is normal. Go straight to the next section.

Pellet and wash the crude RNA

1. Centrifuge at 5,000 × g for 15 minutes at 4 °C. The pellet is transparent and invisible, so handle the tubes gently.
2. Pour off the supernatant.
3. Mix the Tempus Pre-Digestion Wash by gently inverting the bottle 3 to 4 times. Add 4 mLto each tube, recap, and vortex at moderate speed for 5 seconds, enough to draw the liquid about three-quarters of the way up the tube.
4. Centrifuge again at 5,000 × g for 10 minutes at 4 °C.
5. Decant the supernatant carefully. Decanting is recommended because a pipette can take the pellet with it.
6. Drain the tubes inverted on absorbent paper for 2 minutes, then blot the rims.
7. Keep the tubes on ice and move on immediately.

Resuspend the pellet and digest with protease and TURBO DNase

Just before use, prepare the resuspension mixture. Scale it by the number of samples and add 5% extra to cover pipetting loss.

Component	Volume for 1 sample
Tempus Resuspension Solution	117.5 µL
Tempus Proteinase	2.5 µL
Total	120 µL

1. Mix the components by vortexing at moderate speed, spin briefly for 2 to 3 seconds, and keep on ice.
2. Add 120 µL of the mixture to each conical tube, placing it at the bottom of the tube, and recap.
3. Vortex at gentle speed for 5 to 10 seconds to resuspend the pellet.
4. Transfer the full volume to a 1.5 mL microfuge tube or to one well of the 96-well Processing Plate. You can pause here and hold the resuspended pellets on ice for up to 4 hours (cover the plate wells).
5. Add 10 µL of TURBO DNase to each tube or well and mix at vigorous speed for 10 minutes. Do this at 18 to 25 °C. If your lab is colder than 18 °C, use a 25 °C incubator.

Bind the RNA to the beads

1. Vortex the RNA Binding Beads at vigorous speed until fully resuspended.
2. To each tube or well, add 50 µL of Binding Solution Concentrate and 20 µL of the bead suspension. Add the beads before the isopropanol.
3. Mix at moderate speed for 1 minute.
4. Add 200 µL of 100% isopropanol and mix at moderate speed for 3 minutes.
5. Capture the beads on the magnetic stand for about 1 to 3 minutes, until the solution clears.
6. Aspirate and discard all the supernatant without disturbing the beads.
7. Take the tubes or plate off the magnet. This is critical before the next step.

Wash twice with Wash Solution 1

1. Add 150 µL of Wash Solution 1 and mix at moderate speed for 1 minute.
2. Capture the beads for about 1 to 2 minutes and discard the supernatant.
3. Remove from the magnet, then repeat the wash once.

Wash twice with Wash Solution 2

1. Add 150 µL of Wash Solution 2 and mix at moderate speed for 1 minute.
2. Capture the beads for about 1 to 2 minutes and discard the supernatant.
3. Remove from the magnet, then repeat the wash once.

Dry the beads

• Tubes: open them and invert on absorbent paper for 2 minutes at room temperature. If liquid remains, remove it with a fine-tipped pipette without touching the beads, then invert for 1 minute more.
• Plate: shake at vigorous speed for 2 minutes at room temperature. If liquid remains, shake for 2 minutes more.
• Do not over-dry the beads, and make sure no alcohol is left before elution.

Elute the RNA

1. Add 80 µL of Elution Buffer (you can use anywhere from 20 to 80 µL depending on the concentration you want) and mix at vigorous speed for 4 minutes.
2. Capture the beads for about 1 to 3 minutes.
3. Transfer the supernatant to a nuclease-free tube without disturbing the beads. Keep it, because the purified RNA is in this supernatant.
4. Store on ice for immediate use, at −20 °C for up to 6 months, or at −80 °C for the long term.

Checking yield and quality

• Expect an A260/A280 ratio of 1.8 to 2.2. Values outside that range can still work for qRT-PCR.
• For human blood RNA from this kit, a 28S to 18S rRNA ratio of 1.0 or higher is considered acceptable. A RIN from a bioanalyzer is the more reliable measure of integrity.

Troubleshooting

• Low or no yield: check that the tube was filled to the black mark, that the blood was diluted to a final 12 mL with PBS and vortexed for 30 seconds, and that you used the full centrifuge times and speeds in a 50 mL tube in a swing-bucket rotor. Lysate that was exposed to more than 37 °C can be frozen, thawed and repurified. Also check that the beads were fully resuspended, were never frozen, and were added before the isopropanol.
• Lost pellet or beads: decant slowly rather than pipetting off the pellet, and aspirate slowly with the tip away from the captured beads.
• Beads clump during elution: warm the plate or tubes at 70 °C for 5 minutes and repeat the 4-minute mix. Next time, preheat the Elution Buffer to 70 to 80 °C. If the beads were over-dried, extend the elution mixing to 10 minutes.
• Bead carryover (light brown eluate): capture again on the magnet and move the RNA to a fresh tube. A small amount does not inhibit RT or RT-PCR.
• Too much DNA: make sure the digestion ran for the full 10 minutes at vigorous mixing and at 18 to 25 °C, and that the pellet was washed and drained well so stabilizing reagent was not carried over.
• RNA fails downstream: salt or alcohol carryover is the usual cause, so check that the correct alcohol and volume went into each wash concentrate, remove every supernatant fully, and let the beads dry before elution. A green or reddish eluate signals inhibitors, so spin it at 16,000 × g for 3 minutes and move the clear fraction to a new tube.

Source: Applied Biosystems, MagMAX for Stabilized Blood Tubes RNA Isolation Kit (Compatible with Tempus Blood RNA Tubes) Protocol, Part No. 4452007, Rev. A (2010).

Tempus Spin RNA Isolation Kit Protocol for Whole Blood RNA

 

Tempus Spin RNA Isolation Kit Protocol (Cat. 4380204)

This is the manufacturer's protocol for purifying RNA from blood collected in Tempus Blood RNA Tubes, paraphrased from the Thermo Fisher user guide (Pub. 4379232, Rev. E).

Before you start

• The protocol is designed for 3 mL of blood in a Tempus Blood RNA Tube, which contains 6 mL of stabilizing reagent.
• Immediately after collection, shake the tube vigorously or vortex it for 10 seconds. Poor mixing leads to microclots that can clog the filter.
• Stabilized blood keeps for up to 5 days at room temperature and 7 days at 4 °C, and it can be frozen at −20 °C or −80 °C.
• Keep samples on ice as much as possible during processing.

Isolate the RNA

1. If the tube is frozen, thaw it at room temperature (18 to 25 °C).
2. Pour the contents into a clean 50 mL conical tube.
3. Add 3 mL of Ca²⁺/Mg²⁺-free 1× PBS to bring the total to 12 mL. If you collected less than 3 mL of blood, top up with PBS to 12 mL. Skipping this dilution gives much lower yields.
4. Cap the tube tightly and vortex at maximum speed for 30 seconds. Less than 30 seconds can clog the column, and a layer of froth is normal.
5. Centrifuge at 4 °C at 3,000 × g for 30 minutes.
6. Carefully pour off the supernatant. The RNA pellet is transparent and invisible, so take care not to dislodge it.
7. Leave the tube inverted on absorbent paper for 1 to 2 minutes, then blot the rim.
8. Add 400 µL of RNA Purification Resuspension Solution, placing the pipette tip at the bottom of the tube so you do not wash blood residue down the walls. Vortex briefly to resuspend the pellet. The resuspended RNA can wait on ice.

Purify the RNA

Unless stated otherwise, every spin is 30 seconds at 16,000 × g.

1. Place a labeled purification filter in a waste collection tube and pre-wet the membrane with 100 µL of RNA Purification Wash Solution 1.
2. Load about 400 µL of the resuspended RNA onto the filter and spin.
3. Discard the flow-through, add 500 µL of Wash Solution 1, and spin.
4. Discard the flow-through, add 500 µL of Wash Solution 2, and spin.
5. Discard the flow-through and repeat the Wash Solution 2 step once more (500 µL, 30 seconds).
6. Discard the flow-through and spin for 30 seconds to dry the membrane.
7. Move the filter to a new labeled collection tube and add 100 µL of Nucleic Acid Purification Elution Solution. Incubate the whole tube at 70 °C for 2 minutes, then spin for 30 seconds.
8. Load about 100 µL of that eluate back onto the filter and spin for 2 minutes at 16,000 to 18,000 × g. No incubation is needed.
9. Discard the filter. Transfer about 90 µL of eluate to a fresh labeled tube, pipetting from the top of the liquid so the bottom 10 µL and any pellet stay undisturbed.
10. Store the RNA at −20 °C, or at −80 °C for the long term.

Troubleshooting

• If the filter stays black or brownish-red, the usual causes are poor mixing at collection or blood residue contaminating the sample. A repeat wash with 500 µL of Wash Solution 1followed by 500 µL of Wash Solution 2 can help.
• Degraded RNA is blamed on residual protein with RNase activity, or on the lysate being exposed to temperatures above 37 °C. For the first cause, add more wash steps until the membrane looks white.
• Expected yield is 6 to 25 µg of RNA from 3 mL of blood, with an A260/280 ratio above 1.9.

Source: Applied Biosystems, Tempus Blood RNA Tube and Tempus Spin RNA Isolation Kit User Guide, Pub. 4379232, Rev. E (2018).

SYBR Green I Assay for Malaria Drug Testing: Step by Step

Here is a fact I never get tired of: the cell that malaria lives in has thrown away its own DNA. Mature human red blood cells have no nucleus and no genome, so there is nothing in them to stain. That means any DNA you find in a Plasmodium falciparum culture belongs to the parasite. And what does that allow us to do? If we can make DNA glow, we can count parasites, and if we can count parasites, we can ask how much of a drug it takes to stop them.

For a long time, answering that question meant working with radioactivity. Labs fed parasites a radiolabelled substrate and measured how much they took up, which demanded special equipment, safety procedures and a lot of patience. Then, in 2004, Smilkstein and colleagues reported a simple, inexpensive, one-step alternative built on SYBR Green I, a dye whose fluorescence rises sharply when it binds double-stranded DNA (Smilkstein et al., 2004). No isotopes, no wash steps. Just a plate reader. 

In this post I walk through the whole assay, from the drug dilution plate to the IC50, using the layout from my video. It is a 96-well format, so it scales nicely from a single compound to a full screen.

What the assay actually measures

SYBR Green I barely fluoresces on its own. Slip it between the strands of DNA and it lights up, roughly a thousand (1000) times brighter. After 72 hours of growth, you lyse the cells, add the dye, and read the plate. More parasites mean more DNA, which means more green. Add an effective drug and the signal falls. Plot fluorescence against the logarithm of the drug concentration, fit a curve, and the midpoint is your IC50.

One small note on timing. The original paper measured growth after 48 hours, while many labs now use 72 hours so that the parasites go through more than one cycle and slower-acting drugs have a fair chance to show an effect. Pick one, and keep it identical across experiments.

Step 1: Build the drug plate

Prepare your drugs at twice the final concentration in complete medium, because they will be diluted one to one when the parasites go in. Dissolve compounds in DMSO if needed, and keep the final DMSO at or below 0.2% in every well, including the controls.

Add 50 µL of complete medium to the wells that will carry the lower concentrations. Put 100 µL of the top 2× concentration in the first drug column, then transfer 50 µL to the next column, mix, and carry on across the plate, discarding the last 50 µL. Every drug well now holds 50 µL, and each column is half the concentration of the one before. That is your two-fold serial dilution.

Then plan the controls. Edge wells evaporate faster and give odd readings, so fill them with 100 µL of water or PBS and do not use them for parasites. In my layout, columns 2 to 10 of rows B to D hold the test compound, rows E to G hold chloroquine as a reference drug, and column 11 holds the controls: no-drug parasites for 100% growth, and uninfected red cells for background.

Step 2: Prepare the parasite suspension

Hematocrit is the percentage of the culture volume that is packed red cells. We want 2% in the well. Because the parasites are mixed one to one with the drug, you prepare the suspension at 4% hematocrit and the well ends up at 2% in a total volume of 100 µL.

Do you need to add uninfected blood? Only if your culture is above the parasitaemia you want. Parasitaemia is the fraction of red cells that are infected, so the infected cells are already part of the hematocrit and are not counted twice. If your culture is already at 0.5% to 1%, spin it at 500 × g for 5 minutes, remove the supernatant, and use the pellet as your packed cells. If it is higher, dilute the pellet with uninfected red cells until you reach the target parasitaemia.

Here is a bulk mix for one plate, with a little extra for pipetting losses:

Component	Volume (5.5 mL batch at 4% hematocrit)
Packed infected red cells (pellet)	220 µL
Complete medium	5.28 mL

Make a small separate batch of uninfected red cells at 4% hematocrit for the background wells, from the same donor and batch you used to dilute the parasites. That way the background matches the cells in the rest of the plate.

Step 3: Combine and incubate

Add 50 µL of the parasite suspension to each drug well and to the no-drug wells, and 50 µL of the uninfected suspension to the background wells. Each well now holds 100 µL at 2% hematocrit. Place the plate in a humid chamber, flush it with a low-oxygen gas mixture (5% O₂, 5% CO₂, 90% N₂), seal it, and incubate at 37 °C for 72 hours. This is a good time to go and do something else. The plate does not need you.

Step 4: Lyse and stain

The lysis buffer does two jobs. Saponin and Triton X-100 open the red cells and the parasite membranes, and EDTA keeps nucleases quiet, so that the dye can reach the DNA. Prepare it in advance, and add the dye only on the day.

Component	Final concentration
Tris, pH 7.5	20 mM
EDTA	5 mM
Saponin	0.008% (w/v)
Triton X-100	0.08% (v/v)
SYBR Green I (10,000× stock)	0.2 µL per mL of lysis buffer (about 2×)

This is the recipe from the original paper, and it works well for most labs (Smilkstein et al., 2004). SYBR Green I is light sensitive, so protect the buffer with foil. Add 100 µL to every well and mix until no visible red cell sediment remains. Then wrap the plate in foil and leave it for 1 hour at room temperature in the dark.

Step 5: Read the plate

Use a fluorescence plate reader with excitation at about 485 nm and emission at about 530 nm. Black plates reduce crosstalk between wells. After three days of waiting, the reading takes about a minute, which feels almost rude.

Step 6: From fluorescence to IC50

First, subtract the mean of the uninfected red cell wells from every reading. Second, set the mean of the no-drug wells to 100% growth. Third, plot the percentage growth against the log of the drug concentration and fit a sigmoidal dose-response curve with nonlinear regression, in GraphPad Prism or a similar program. The concentration at the midpoint of the curve is your IC50.

Run each concentration in triplicate, and repeat the whole experiment on at least two or three separate days. Technical replicates tell you how well you pipette, while independent experiments tell you whether the result is real. And check your reference drug: if the chloroquine IC50 for a well-characterised strain is far from the values reported in the literature, the assay is telling you something, and it is not about your compound.

Things that bite you

• White blood cells. They carry DNA too. This is no problem for cultured parasites, but it matters for fresh patient blood unless you remove the leukocytes first.
• DNA-binding compounds. A drug that grabs DNA competes with the dye and can fake a result, and a coloured compound can quench the signal.
• Slow-acting drugs. Some compounds need a longer exposure than 72 hours to show their full effect.
• Edge wells. Fill them with water or PBS, not parasites.
• Light and age. Protect the dye from light and make the final buffer fresh.

Watch the video

Prefer to see it done? I walk through the whole assay, step by step, in the video below, including the plate layout and the dose-response curve.

Video: SYBR Green I Assay: Malaria Drug Testing Step by Step

Related Content

• Hematocrit calculation for passaging P. falciparum
• The P. falciparum life cycle
• Giemsa stain: morphology and history
• Red blood cell isolation protocol

References

Smilkstein, M., Sriwilaijaroen, N., Kelly, J. X., Wilairat, P., & Riscoe, M. (2004). Simple and inexpensive fluorescence-based technique for high-throughput antimalarial drug screening. Antimicrobial Agents and Chemotherapy, 48(5), 1803-1806. https://doi.org/10.1128/AAC.48.5.1803-1806.2004

RNA Extraction Using Norgen Biotek: A Step-by-Step Laboratory Protocol

If you are extracting RNA from blood in Tempus tubes and you want every size of RNA, including the small species, the Preserved Blood RNA Purification Kit I from Norgen Biotek is worth a close look. It is a spin-column kit made specifically for Tempus tubes, and it is the quickest of the three Tempus-compatible kits to run.

The workflow will feel familiar. You dilute the stabilised blood, pellet the RNA, dissolve the pellet in a guanidinium-based lysis solution, add ethanol & bind to a column. 

Norgen's resin binds RNA depending on the ionic conditions, so proteins pass through or stay on top of the resin, with no phenol or chloroform. The kit also offers an optional on-column DNase step, which is handy when genomic DNA matters.

So how does it run on the bench? Here is the full procedure, step by step.

You'll need

Kit

Norgen Biotek Preserved Blood RNA Purification Kit I, for use with Tempus Blood RNA Tubes (product no. 43400, 50 preps). All components are stored at room temperature and stay stable for at least a year unopened.

• Tempus Blood RNA Tube Diluent
• Lysis Solution
• Wash Solution (supplied as a concentrate)
• Elution Solution
• Mini spin columns
• Collection tubes
• Elution tubes (1.7 mL)

You'll also need

• Blood collected in Tempus Blood RNA Tubes
• 95 to 100% ethanol (and 95% ethanol for the wash solution)
• Sterile 50 mL conical tubes
• Swing-bucket centrifuge that can run at 4 °C
• Benchtop microcentrifuge capable of 14,000 × g, ideally with variable speed
• Vortexer
• Optional: Norgen RNase-Free DNase I Kit (product no. 25710) for the on-column DNase step
• Gloves, lab coat and eye protection, because this is human blood

Part 1: Prepare the Wash Solution

Do this before starting the samples. Add 50 mL of 95% ethanol to the bottle of concentrated Wash Solution. This gives a final volume of 72 mL. Tick the box on the label to show that the ethanol has been added.

Make sure all solutions are at room temperature before use, & work quickly through the whole procedure, because speed protects the RNA.

Part 2: Prepare the lysate

Step 1: Transfer to a conical tube

Pour the entire contents of the Tempus tube into a new 50 mL conical tube.

Step 2: Dilute with Diluent

Add 3 mL of Tempus Blood RNA Tube Diluent, or enough to bring the volume to 12 mL. Close the tube tightly & vortex vigorously for 30 seconds.

Step 3: Pellet

Centrifuge at 4 °C at 3,000 to 5,000 × g (at least 4,500 rpm in a swing-bucket rotor of the Beckman JB-6 type) for 30 minutes. The RNA pellet is transparent and invisible.

Step 4: Drain

Discard the supernatant, then leave the tube inverted on a paper towel for 2 minutes to remove excess liquid.

Step 5: Lyse & add ethanol

Add 600 µL of Lysis Solution to the pellet & vortex for a few seconds to resuspend it. Then add 300 µL of 95 to 100% ethanol & vortex briefly to mix.

Part 3: Bind the RNA to the column

From here on, all centrifugation is done at room temperature in a microcentrifuge at 14,000 × g, unless stated otherwise. The manufacturer warns that temperatures below 15 °C can form precipitates that clog the column.

Step 6: Load

Assemble a column with a collection tube. Apply up to 600 µL of the lysate and ethanol mixture and centrifuge for 1 minute at least 3,500 × g (about 6,000 rpm).

Check that the whole volume has passed through. If not, spin for 1 additional minute at 14,000 × g.

Step 7: Repeat if needed

Discard the flow-through, reassemble the column and collection tube, and repeat Step 6 until all of the lysate has been loaded. With 900 µL of lysate in total, you need two loads.

Part 4: Optional on-column DNase treatment

The kit already gives low genomic DNA carry-over, but the manufacturer provides this step for sensitive downstream work. It is done right after binding, so decide before you start (required if you are doing RNA-Seq). It is particularly worth considering for parasite qRT-PCR.

Step 8: Prepare the DNase

For each column, prepare a mix of 15 µL of RNase-free DNase I and 100 µL of Enzyme Incubation Buffer from Norgen's DNase kit, then mix by inverting. Do not vortex. Apply 100 µL of the solution to each column in Step 10.

Step 9: Pre-wash

Apply 400 µL of Wash Solution to the column and centrifuge for 2 minutes. Discard the flow-through and reassemble the column.

Step 10: Apply and incubate

Apply 100 µL of the DNase solution to the column and centrifuge for 1 minute at 14,000 × g. Pipette the flow-through back onto the top of the column, which is needed for full DNase activity and the best recovery of small RNAs. Incubate the column assembly at 25 to 30 °C for 15 minutes.

Then go straight to the second wash in Part 5 (Wash 2) without any further spin.

Part 5: Wash

Step 11: Wash three times

Add 400 µL of Wash Solution to the column and centrifuge for 1 minute. Check that the full volume passed through, discard the flow-through and reassemble the column. Repeat for a second and third wash. Three washes are needed to clear the salt from the binding step.

Step 12: Dry the resin

Spin the column for 2 minutes to dry the resin thoroughly. This removes traces of ethanol that can interfere with downstream applications. Discard the collection tube.

Wash	Solution	Volume (µL)	Spin
1	Wash Solution	400	1 minute at 14,000 × g
2	Wash Solution	400	1 minute at 14,000 × g
3	Wash Solution	400	1 minute at 14,000 × g
Dry	None	None	2 minutes at 14,000 × g

Part 6: Elute

Step 13: Elute

Place the column in a fresh 1.7 mL elution tube. Add 50 µL of Elution Solution. Centrifuge for 2 minutes at 200 × g (about 2,000 rpm), followed by 1 minute at 14,000 × g. Note the volume eluted. If the full 50 µL has not come through, spin at 14,000 × g for 1 additional minute.

Step 14: Second elution (recommended)

For maximum recovery, repeat Step 13 with another 50 µL of Elution Solution into a separate tube. Keep the two eluates apart, since the second is more dilute.

Store the RNA at −20 °C for a few days, or at −70 °C for the long term.

How do you know it worked?

The manufacturer quotes an average of 5 to 25 µg of RNA from 3 mL of preserved human blood, with RNA of all sizes, including species below 200 nucleotides. One column binds a maximum of 50 µg. With parasitised patient blood, quantify with a fluorometric method such as Qubit or RiboGreen and check the traces on a bioanalyser, since the RNA is a mix of host and parasite RNA.

Troubleshooting tips

• Poor recovery: check that the full volume of Lysis Solution was used, that ethanol was added to both the lysate and the Wash Solution, and that you used the supplied Elution Solution.
• Clogged column: usually caused by incomplete solubilisation of the blood, or by a centrifuge that is cooler than 15 °C during the column spins.
• Degraded RNA: RNase contamination, a procedure that took too long, or poor storage of the purified RNA.
• Poor performance downstream: carry-over of salt (do all three washes) or ethanol (do not skip the dry spin).
• Genomic DNA contamination: use the on-column DNase step, or treat the eluted RNA with RNase-free DNase I.

Reference

Norgen Biotek Corp. (2015). Preserved Blood RNA Purification Kit I (for use with Tempus Blood RNA Tubes) product insert (Product No. 43400, PI43400-3).