Question

You are in a laboratory attempting to identify a genetic defect responsible for a disease. You believe you have located the gene that results in a faulty protein—but you aren't sure!

How could you be sure you have located a section of DNA that encodes for a protein?

Write down the sequence to see what amino acids might be linked together.

Allow the DNA to be transcribed to RNA and see what protein results.

Compare the faulty protein to the DNA.

Answer 1

Short answer: do both in silico prediction and experimental validation. The three ideas you listed are all useful steps, but none alone is definitive.

A practical roadmap

1) In silico checks (quick, inexpensive)
- Identify the coding region/ORF:
- If you have genomic DNA, first map exons (genome browsers, gene models) or PCR/sequence cDNA — genomic sequence usually contains introns, so don’t naively translate genomic DNA.
- Use ORF-finding/translation tools (NCBI ORF Finder, ExPASy Translate) to find start codon(s) and reading frames and predict the amino acid sequence.
- Check for hallmarks of a protein-coding gene: conserved open reading frame, start codon (ATG, good Kozak context in eukaryotes), in-frame stop codon, splice sites, promoter elements, codon usage, conservation across species.
- BLAST the predicted protein against protein databases; presence of known domains (Pfam, InterPro) increases confidence.

2) Verify transcription (shows RNA is produced)
- RT-PCR or qRT-PCR for the transcript from relevant tissue/cells.
- RNA-seq data or Northern blot to demonstrate transcript length and expression.
- Sequence the cDNA to confirm exon usage and reading frame (rules out splicing errors).

3) Verify translation (shows protein is made)
- Western blot with a specific antibody (size consistent with predicted protein).
- Mass spectrometry (peptide mapping) to confirm sequence/peptides from the predicted protein.
- Ribosome profiling (Ribo-seq) to show ribosomes engage the ORF.
- In vitro translation or heterologous expression (clone the cDNA into an expression vector, express in cells or cell-free system) and detect the produced protein.

4) Link genotype to phenotype (show the variant causes the faulty protein / dysfunction)
- Compare the predicted amino-acid change from the DNA variant (missense, nonsense, frameshift, splice-site) to the observed protein (Western, MS). Confirm that the variant produces the altered protein predicted from the DNA/cDNA sequence.
- cDNA sequencing to see whether the mutation is present in the transcript and whether it causes exon skipping, intron retention or frameshift.
- Functional assays or complementation: express wild-type vs mutant cDNA in cells (or model organism) and test the protein’s function; rescue experiments can be decisive.
- Rule out nonsense-mediated decay for premature stops (measure RNA levels).

Practical notes & pitfalls
- Don’t translate raw genomic DNA without accounting for introns—use cDNA or annotated exon sequence for accurate translation.
- Alternative splicing can produce multiple protein isoforms; check tissue-specific transcripts.
- Post-translational modifications won’t be predicted from the DNA sequence, but can affect size/function.
- Some ORFs are small or non-coding RNAs can have ORF-like stretches but not produce functional protein—use experimental evidence of translation.

If you want, paste the DNA sequence (or indicate whether it’s genomic vs cDNA), and I can:
- find ORFs and translate to amino-acid sequence,
- predict the impact of any variant,
- suggest the best experiments to validate it in the lab.