Biology IA · D · Continuity and change · topic 21 of 25

How often does a mutation change a protein? · IB Biology IA idea: guide, sample graph and step-by-step plan

no lab, but a LOT of thinking. nice.

Change one letter of a gene and the protein may not notice, may swap one amino acid, or may stop dead. Count how often, for every possible change.

Difficulty●●○ Medium
KitComputer only: NCBI sequences, a spreadsheet or a short script
TimeAbout 10 hours
FitsSL and HL
  • ✓ you like patterns and a bit of code or spreadsheet logic
  • ✓ you want a biology IA with no lab and no mess
  • ✓ you enjoy testing every possibility, not just a few

The big idea

← test every possibility

The genetic code is redundant, so many single-base changes are silent. Others swap an amino acid (missense), and a few create a stop codon (nonsense). Which one is likely depends on the codons a gene actually uses, and codon use is linked to GC content. You take real genes from a database, test every possible single-base change in each, and compare genes of different GC content. You are not predicting a lab result: you are working out what the code makes inevitable, then asking whether GC content shifts the balance.

Make it yours: pick one twist

Copying a classic won't show your choices. Pick one of these (or invent your own) and your IA becomes yours.

Swipe for more twists →

Weight the changes

Real mutations are not all equally likely: transitions (A↔G, C↔T) are commoner than transversions. Give them a documented weighting.

→ Your percentages become weighted; you need a cited ratio, and you can compare weighted and unweighted results.
GC at the third position

Use GC content of the third codon base (GC3) instead of the whole gene.

→ Your independent variable matches where silent changes happen; you must explain why this might predict the result better.
Compare different organisms

Choose genes from bacteria, a plant and a mammal, with different average GC content.

→ Species becomes a confounding factor; you must say how you choose genes and why different species are comparable.
Look at position

Compare the first, second and third base of each codon rather than lumping them.

→ You get a second analysis (where mutations matter most), and a clear link back to the code's structure.
Compare gene families

Use a set of genes with a shared function (such as ribosomal proteins) against a random set.

→ Your independent variable becomes gene type; you discuss whether function constrains the result.
Change the context

Link it to a human disease gene, antibiotic resistance, or why viruses tolerate mutation.

→ Gives the introduction a real purpose; the counting method stays the same.

Your Biology IA research question: run the check

Write your own question first. Then make sure it ticks every box:

  • Names the independent variable (how GC content is defined and calculated) and the range you will cover.
  • Names the dependent variable as a percentage of all possible single-base changes, and says which types it is split into.
  • Says which genes, from which database and organism, and how many genes you will use at each GC level.
  • Says how you choose genes (random? a filter on length?) so the choice is not your own bias.
  • States the genetic code table you use (standard or not) and how you treat start and stop codons.
  • Is small enough to be checked by hand on a short sequence and run on all your genes in about 10 hours.

Get your RQ checked · S$29

The set-up

No apparatus here, so the picture is the method. Draw yours with your own database, your spreadsheet or script and your choices.

gene coding sequence (NCBI)cut into codonsN1N2N3silentsame amino acidmissensedifferent amino acidnonsensenew stop codontally every codon of the gene: % silent, % missense, % nonsensereading frame!starts at ATG3 positions × 3 new bases = 9changes per codontest it by hand on a 30-base sequence first

How your Biology IA could be laid out

The usual order for a topic like this, what each part needs to show, and roughly how much of your 3,000 words it takes. The words are all yours.

Research question and context~250
Your question; why mutation types matter; why GC content might shift the balance, and whether you can justify a direction at all.
Background biology~450
The genetic code, its redundancy, silent, missense and nonsense changes, and what GC content is. Link to transcription and translation.
Variablestable
Independent, dependent and controlled variables, including how gene choice is controlled.
Method~550
Database and how you searched; how genes were chosen and filtered; how codons are read and the frame checked; how every change is generated and classified; how you checked the code on a short sequence; how GC content was calculated.
Safety, ethics, environment~40
Short: screen time, and citing data sources and accession numbers.
Raw datatable
For each gene: accession number, length, GC content, count of each type of change. A short extract in the report, the rest in an appendix.
Processing~300
From counts to percentages per gene, then averages and spread per GC group; one hand-checked example.
Graphs and analysis~400
Percentage of each type against GC content with error bars, each gene as a point; a correlation or regression suited to the design, with conditions checked.
Conclusion~350
Answer with values and spread; say whether differences are bigger than the gene-to-gene spread; limit the claim to the genes and range you used.
Evaluation~550, the biggest part!
Weaknesses ranked by their effect on the percentages: gene choice, frame errors, unweighted changes, small number of genes, the GC range actually available.
Referencesnot counted
Database, accession numbers, genetic code table and any theory sources.

What your Biology IA graph could look like

a flat result is still a result!

Made-up data, to show you the form of a strong graph, not the answer. Only the no-effect line is drawn: whether there is a slope, and which way it goes, is for your data to show. example only

303540455055606570202122232425262728Gene GC content / % (one point per GC group of genes)Silent changes / % of all possible single-base changes (mean ± SD)no-effect line (what you test against)mean of genes in each GC group, with SDa slope up OR down only countsif it beats the gene-to-gene spreadILLUSTRATIVE · MADE-UP DATA
  • Each point is a group of genes, and the bars are the spread between genes. The unit of replication is the gene, not the codon.
  • Only a flat no-effect line is drawn. Fit your own line and report its slope with an uncertainty; do not decide the direction before you have the data.
  • The axis starts at 20, not zero, so a small effect is visible. Say so in the text.
  • Plot every gene as its own point as well, if you can: it shows the spread and any outlier gene.
  • Another graph to consider: the three percentages (silent, missense, nonsense) stacked for each GC group.

Your game plan

the hand check is your trial run

Tick steps off as you go. We'll remember on this device.

0 of 10 done

  1. Read and predictLearn the genetic code and why it is redundant. Write down what you think GC content will do to the silent percentage, and why.
  2. Do it by hand on a short sequenceTake a 30-base coding sequence. List all changes in each codon and classify them on paper. This is your trial run, and the check for your sheet or script.
  3. Build the counterIn a spreadsheet or short script, split a sequence into codons, make the nine changes, translate and classify them. Test it on the 30-base sequence until it matches your hand count.
  4. Choose genesDecide where the genes come from, how many per GC group, how you avoid picking the ones you like, and any length filter. Write down why.
  5. Check reading framesCheck each sequence starts at the start codon, has no internal stop and is a multiple of three. Decide what to do with the ones that fail, and record it.
  6. Run every geneRun the counter on all genes. Record accession numbers, length, GC content and counts as you go, so you can trace every number.
  7. ProcessPercentages per gene, then mean and spread per group. Decide how you treat start and stop codons and say why.
  8. Graph and testPlot each type against GC content with error bars. Choose a statistic that fits the data, check its conditions, and say what the result means for the RQ.
  9. Conclude and evaluateAnswer with values and uncertainty, compare with published figures for the genetic code, and rank your weaknesses by their effect on the result.
  10. Check before your teacher's draftRun the checklist below, then get your RQ or draft checked. Your teacher reads one draft only, so make it count.

Biology you'll need

Traps to avoid in this Biology IA

  • A reading-frame error: one base out and every codon after it is wrong, so everything you count is junk.
  • Treating the codons within one gene as replicates. The gene is the replicate; thousands of codons from one gene is still one gene.
  • Assuming that every possible change is equally likely, without saying so or testing a weighting.
  • Including the stop codon at the end of the gene as if it could be mutated like the others.
  • Not hand-checking the counter on a short sequence before trusting it.
  • Choosing genes by hand, so the GC groups differ for a reason other than GC (such as length or function).
  • Not giving accession numbers, so nobody can repeat your data.

More Biology IA ideas like this

Make sure it's yours

This guide shows how a topic like this is usually approached. It is not an IA. Your question, choices, data, analysis and words must be your own: change something, tell your teacher where the idea came from, and follow your school's rules on outside help.

Get your RQ checked · S$29