Your passage number might be affecting your data more than you think❗ Liu et al. (2025) looked at RNA-seq data across passages P3–P39. (a thread)
Wildtype One
@wildtypeone.bsky.social
🧫 Helping researchers diagnose ugly lab data: biology, statistics, or the wrong readout | Join 1,000+ researchers getting weekly notes (it’s free) 👉 wildtypeone.substack.com/about
Before you repeat the experiment, answer this: What kind of problem do you actually have? Ugly data does NOT have one next step. (a thread)
“Everyone does it this way” is not a valid statistical defense. (1/3)
“Do more replicates until it’s significant” is killing your credibility 🧫 (a thread)
26 HARD lab truths biology researchers take years to learn: 1. You don't have to respond to all reviewer comments. Some will harm your project. ... (a thread)
The biggest discount your lab will ever get? (a thread 🔬💰👇)
🧬 DNA found no candidate resistance mechanism in 51% of resistant patients. 📃 mRNA explained only ~40% of protein-abundance variation. 🧱 Protein abundance explained only ~10% of enzyme-activity variation. One omic layer is not the whole truth. Omics need YOUR judgment. — Wildtype One
Don't always trust expiry labels on lab reagents (at least if you're in a rush) Here's proof from a wild (and true) story 👇 (1/5)
It uses the same samples as Western blots. But instead of taking 2 days to reveal 2 proteins... Kinome profiling reveals all 400+ major signaling kinases ...in two hours. (1/5)
Biologists lose some of their work's value when they chase aesthetics: - SEM instead of SD - Dynamite plots everywhere - Not showing sample and effect sizes The biology is in the nuance. 🧬 Catch yourself chasing aesthetics. Show the variability.
The last year of the PhD/postdoc is NOT "just another year." It is when everything concludes. Like the last episode of a series. (a thread)
According to 2025 data, you can cut 60% of your flow failures with 4 fixes only (see thread 👇)
Scan your manuscript—fraud can be unintentional (a thread)
Stop making “clean” figures that hide your data. Bar charts compress ❌ SEM hides variability ❌ One chart ≠ every experiment ❌ Good figures show: • all points (small n) • distribution (large n) • direction (paired data) If your figure looks clean but hides truth, it’s not good science. 📊
13 cases where your experiments disagree & what it means (a thread 👇)
Contamination rarely looks like contamination Here are unusual signs we often ignore 🚩👇 (a thread)
Never buy an antibody because of: - Brand ❌ - Price ❌ - Number of citations ❌ Only buy antibodies if: It has been validated for your exact application and species ✅ When buying antibodies, validation is king. It saves you time, money, and failed experiments.
To all hard-working researchers: Working long hours & weekends does NOT prove that you're serious. No one cares how long you work. Exhausted work = less depth = worse judgment We're paid to use our thinking brains. Brain performance requires recharging. — Wildtype One 🧬
The fastest guide to fix Western blots—8 failures in 30 second Save for emergencies 🚨 (A thread👇)
Stop using SEM because it’s cleaner than SD (see thread)
Western blots are great for focused questions. But are not discovery tools. Too many failure modes. Only semi-quantitative. Technically fragile. Use high-throughput methods for discovery. Use western blots to validate. Don't invert the sequence.
You can remove outliers without feeling guilty—under one condition only: If it's a "T-case." What's a T-case? 🤔 Read today's Wildtype Weekly to find out (in 3 minutes or less 👇 )
You can remove outliers—under one condition
“If you torture the data long enough, it will confess to anything.” — Ronald Coase Dear Researcher, We’ve all had unusual data that: didn’t reproduce “ruined” our figures caused statistics to be “non-...
linkedin.com
Imagine Einstein had an hour to fix your experiment 👴🏻 🧬 (a thread)
If your experiments worked but your figures still feel fragile, it’s often not the biology—it’s the analysis. This webinar walks through the 5 most common statistics mistakes biologists make + how to avoid them No coding. No equations. Just clarity. 👉 Register for the webinar here: wildtypeone.com
Irreproducibility undermines the biggest research breakthroughs And we hear more false data today than ever Below are 7 Method-section phrases that can kill doubt and have people trust your work All in 3 minutes or less—read today's Wildtype Weekly 👇
7 small manuscript lines that kill doubt
“We are drowning in information, while starving for knowledge.” — Sydney Brenner Dear Researcher, This one phrase in a 2026 Method section gave a confidence boost for the whole article: Irreproducibil...
linkedin.com
“Everyone does it this way” is not a valid statistical defense. (a thread)
