Generation STEM
Learning LibraryBiotech LibraryTeens

Sequence analysis: measuring DNA with code

Once DNA is a string, you can measure it. Counting bases and computing GC content are the first tools of bioinformatics — and they're a few lines of Python.

The big idea

Sequence analysis uses string operations to measure a sequence — counting bases, computing GC content, comparing strands.

See it in code

1The basics

Measuring a strand starts with counting. len gives its size, and .count() tallies how many times a single base appears:

python
dna = "ATGGCGCGCTTAA"
print("Length:", len(dna))
print("G count:", dna.count("G"))
print("C count:", dna.count("C"))
Run it — the strand's length and two base counts:
Length: 13
G count: 4
C count: 3

dna.count("G") scans the whole strand for you. These raw counts are the raw material for every measurement that follows.

2A step further

Counts become insight once you turn them into a ratio. Add the G and C counts, divide by the length, and you have the fraction of the strand that's G or C:

python
dna = "ATGGCGCGCTTAA"
gc = dna.count("G") + dna.count("C")
fraction = gc / len(dna)
print("G + C:", gc, "out of", len(dna))
print("GC fraction:", round(fraction, 2))
Run it — the G+C total as a fraction of the whole strand:
G + C: 7 out of 13
GC fraction: 0.54

gc / len(dna) is the whole idea behind GC content. Next we report it the way biologists do — as a percent.

3In our world

Now all four counts together, and GC content — the fraction of G and C — formatted as the percent biologists actually report, because GC-rich regions bind more tightly:

python
dna = "ATGGCGCGCTTAA"

a = dna.count("A")
g = dna.count("G")
c = dna.count("C")
t = dna.count("T")

gc_content = (g + c) / len(dna) * 100
print("Counts -> A:", a, "T:", t, "G:", g, "C:", c)
print(f"GC content: {gc_content:.1f}%")
Run it — base counts and a GC-content percentage:
Counts -> A: 3 T: 3 G: 4 C: 3
GC content: 53.8%

.count and a little arithmetic turned raw sequence into a meaningful statistic. Real genomes are billions of bases long, but the code is identical — that's the power of measuring data instead of eyeballing it.

The same idea, everywhere

Counting and computing ratios over a sequence is a universal analysis move: word frequencies in a document, error rates in a log, the share of a category in a dataset. Bioinformatics scales it up — the same count and percentage over a whole chromosome.

Try it yourself

Compute AT content and confirm it plus GC content equals 100%. Then write a gc(dna) function and run it on several strands to compare which is GC-rich.

The common mistake

Dividing by the wrong total, or forgetting sequences can contain unexpected letters (like N for 'unknown'). Always divide by len(dna), and be ready for real data that isn't a clean four-letter string.

What it unlocks

Analysis builds on string methods, operators, and f-strings, and extends to comparing orthologs across species.

Related concepts