Skip to content
~/sultan
Chapter 2 crest
Chapter 2 · § 2.1 · Recipe

Exome beats Whole Genome

A contrarian sequencing decision, argued carefully.

By · Updated March 2026 · Last reviewed March 2026 · ~6 min

Problem

You're building the foundation of a personal longevity program. Three genetic testing options are in front of you — microarray, exome NGS, and whole genome sequencing — each at a different price point. Which one actually pays back?

Solution

Get exome. Once, ever. Skip the other two.

Test Cost Grade Verdict
Microarray
23andMe-style chip		 $100–$200 C Coverage insufficient, datasets dated. Useful for ancestry, not longevity.
Exome NGS
~20,000 coding genes		 $300–$800 A The right default. Best information density per dollar.
Whole Genome
~3 billion bp		 $300–$1,500 C Error-prone at consumer 30× coverage. Marginal info over exome.
✦ Tip Ask the lab for the raw VCF file in addition to the interpreted PDF report. A second-pass interpretation down the road will need it.

Discussion

At consumer-tier coverage (typically 30×), whole genome sequencing has higher real-world error rates than advertised. When I ran WGS and exome on the same participants in Trial 2, the exome data was cleaner, required less re-interpretation, and produced practically identical recommendations — at a fraction of the cost.

The case for WGS rests on non-coding variation mattering more than exome captures. That is genuinely true for rare-disease workups and some polygenic risk scores. It is mostly not true for the decisions a consumer longevity program actually makes. The variants that drive drug metabolism, nutrient processing, and inflammation pathway function are protein-coding. Exome gets them.

△ Warning Most "consumer" WGS products ship at 30× coverage. If you genuinely need genome-level data — a rare-disease workup, a structural-variant question — demand ≥60× coverage and work with a clinical geneticist. The consumer tier is not the same thing.

If and when WGS is both cheaper than exome and reliably interpretable at consumer coverage, I will change my mind. For now, exome is the right default.

ℹ Note Exome NGS produced roughly 4,000 usable features per participant in my cohorts. One of those features — a specific biotin-processing variant — led to the case-study participant who lost 45+ pounds and dropped 9 years of biological age (See § 6.4 · The Biotin Case).

See Also

❦ ❦ ❦