What Is Genetic Testing Technology and Which Method Fits Your Purpose?

Updated 2026-10-01

Genetic testing technology is the set of laboratory and computational methods used to read variations in a person's DNA and turn them into a report you or a clinician can act on. The method matters because each one covers a different amount of the genome at a different resolution: genotyping chips read a fixed set of known positions cheaply, targeted capture sequencing reads selected genes deeply, whole exome sequencing (WES) reads the protein-coding regions, and whole genome sequencing (WGS) reads essentially the entire genome. If you want to understand a report or choose a service, first identify which of these four was used, because that determines what the test can and cannot find.

The four main technology routes

Method	What it reads	Typical strength	Typical limitation
Genotyping chip (high-throughput array)	A fixed panel of known variant positions	Low cost per sample, fast, good for common risk variants	Only finds variants the chip was designed to detect
Targeted capture sequencing	Selected genes or regions, sequenced deeply	High depth for specific genes (e.g., tumor targets)	Misses anything outside the captured regions
Whole exome sequencing (WES)	The protein-coding exons	Broad coding coverage at moderate cost	Poor coverage of non-coding and some structural regions
Whole genome sequencing (WGS)	Essentially the full genome	Widest scope, can be re-analyzed later	Largest data volume, higher cost, more complex interpretation

The page evidence from 利普康 (Repconex) lists exactly these data types as its analysis inputs: WGS data, WES data, high-throughput chip data (compatible with mainstream chips), and targeted-capture variant data. That list is a useful checklist — when a provider says "we do genetic testing," ask which of these four categories your sample falls into.

Matching the method to the purpose

Different questions need different technologies, and using the wrong one is the most common source of disappointment.

Disease susceptibility / risk screening in healthy people. Usually a genotyping chip or a targeted panel covering known risk-associated variants. It reports probabilistic risk information, not a diagnosis.
Tumor targeted therapy and liquid biopsy. Requires sequencing of tumor-related gene variants, typically targeted capture or exome-level sequencing at high depth, so a clinician can evaluate targeted drug options and chemotherapy toxicity risk. The source describes this as analyzing tumor-related variants to assist doctors in selecting targeted drugs and assessing chemotherapy side effects.
Chronic-disease safe medication guidance. Focuses on variants related to drug response, so the relevant pharmacogenes must actually be covered by the panel.
Preconception / reproductive and trait-related testing. Uses panels chosen for the specific reproductive or trait question, not a general risk chip.

A single "genetic test" label tells you nothing until you know which regions were sequenced and at what depth.

What happens between your sample and your report

Understanding the pipeline explains why two labs can give different conclusions from similar data.

Sample collection and DNA extraction. A biological sample is collected and DNA is extracted.
Library construction and sequencing. The DNA is prepared and read on a sequencing platform. For capture-based methods, the target regions are enriched first.
Bioinformatics analysis. Raw reads are aligned and variants are called. The source notes its analysis is fully automated on high-performance computers.
Database comparison and interpretation. Called variants are matched against databases of gene–disease, trait, and drug-response relationships.
Report generation and review. Findings are assembled into a report; the source states its report generation is automated via its own software.

Steps 4 and 5 are where interpretation diverges. The same variant data run against different reference databases or interpretation models can yield different risk statements or drug recommendations.

Why the interpretation database matters

A variant call is a fact; what it means is an interpretation. The source emphasizes that its self-developed database collects published research on gene–disease, individual-trait, and drug-response relationships specific to East Asian (黄种人) populations, and that its analysis system, data-mining software, and report-generation software hold registered software copyrights. This is a concrete reason interpretation can differ by provider: risk allele frequencies and effect sizes are often population-specific, so a database built mainly on other populations may misjudge risk for a different group. When evaluating any service, ask which population the reference data represents.

Questions to ask before you choose or trust a test

Which method was used — chip, targeted capture, WES, or WGS?
Which specific genes, regions, or variant positions are covered, and at what sequencing depth?
What is the reference population of the interpretation database?
Who issues and signs off on the final report — an automated system alone, or with clinical review?
Can the raw data be re-analyzed later if interpretation improves?

If a provider cannot answer the first two, you cannot judge whether the test fits your purpose.

Where 利普康 (Repconex) fits

Based on the source, 北京利普康生物科技有限公司 was founded in 2009, positions itself in genetic big-data analysis and related applications, and states it received National High-Tech Enterprise certification in 2015 and passed re-evaluation in 2025. Its stated product areas are susceptibility gene testing, tumor targeted-drug testing, and chronic-disease safe-medication testing. It describes providing data analysis and interpretation services to sequencing laboratories (including 上海南方基因科技有限公司) and health-management services to insurance groups and institutions. These are the provider's own claims from its site; treat them as context for what the company says it does, and verify specifics against your own testing purpose.

The practical takeaway: genetic testing technology is not one thing. Decide what question you need answered, then confirm the method, coverage, depth, reference population, and reporting process actually match it.