Evidence level: B
基因风险评分已确证能提升疾病预测,皮纹是另一条\u201c出生即定型\u201d的信息通道——两条通道联合,增量价值有多大?
正文
先看一个反常现象。皮纹(指纹、掌纹的脊线结构)预测疾病,文献结论忽左忽右:精神分裂症的 meta 分析显示患者总指脊数与 a-b 脊数显著低于对照(小但可靠的效应),高血压的系统综述也报告患者斗型纹频率与总脊线数更高;可另一项 300 人的前瞻研究却报告皮纹与 2 型糖尿病、高血压完全无统计关联。同一类指标,为什么有时灵、有时不灵?
答案可能藏在预测医学最近十年的一个热点里:多基因风险评分(PRS)。PRS 的本质,是把成千上万个微效风险位点加权求和,得到一个描述个体\u201c多基因遗传负荷\u201d的分数。芬兰 FinnGen 生物库(13.5 万人)的研究显示,高 PRS 者终生风险比平均 PRS 者高 21%–38%,发病平均提前 4–9 年,并在糖尿病、房颤、乳腺癌、前列腺癌上显著改善风险分层;英国生物库 32 种复杂疾病的分析则给出更系统的数字:PRS 加入临床预测模型后平均提升约 6% 的预测准确性,心衰更是从 C-index 0.546 跃升到 0.751(+37.5%),且效应在早发人群中更强。一句话:PRS 的增量价值已被大规模生物库证实,虽然\u201c增量\u201d多数时候是\u201c小幅但真实\u201d。
PRS 有一个被反复强调的卖点:出生即可量化、终身不变、先于一切临床风险因素。这一点和皮纹几乎一模一样——皮纹在胚胎第 10–24 周定型,终身不变,遗传决定系数超过 0.80,同样是一种\u201c出生即定\u201d的个体标记。但两者有一个关键差异:PRS 只捕捉\u201cDNA 序列里写死的遗传负荷\u201d,而皮纹作为形态学印记,额外捕捉了胚胎期基因与宫内环境的交互结果——孕早期缺氧、营养、激素波动都会改写脊线发育轨迹。换句话说,皮纹可能携带\u201cPRS 读不到\u201d的非遗传但出生定型的印记。
这正是联合假说的起点。如果皮纹与 PRS 是两条部分独立的信息通道,那么联合模型的增量价值就取决于两者的相关程度:完全相关则无增量,完全不相关则增量最大。现实应在两者之间——皮纹与疾病风险基因共享部分发育通路(这也是皮纹 GWAS 位点富集肢体发育通路的原因),但宫内环境印记的部分大概率是独立的。单用皮纹预测疾病之所以忽左忽右,可能正是因为它的信号弱、被样本量和小效应淹没;而 PRS 提供了稳定的遗传主干,皮纹则可能在主干之外补充\u201c环境修饰\u201d的分支信息。
由此推出三个可检验命题。命题一:在精神分裂、高血压、冠心病队列中,\u201c皮纹+PRS\u201d联合模型相对单用 PRS 的 ΔAUC 与净重分类改善是否显著为正——这是增量价值的直接检验。命题二:低 PRS 但皮纹异常者(如 a-b 脊数偏低、波动性不对称升高)的疾病风险是否仍高于低 PRS 且皮纹正常者——这检验皮纹是否携带独立于遗传负荷的宫内环境印记。命题三:皮纹-疾病关联中多大比例由 PRS 介导、多大比例独立于 PRS——用中介分析把\u201c遗传共因\u201d与\u201c环境印记\u201d两条路径拆开。
证据与局限
已验证: PRS 在临床风险因素之上提供真实增量(FinnGen 5 病种、UK Biobank 32 病种,A 级);皮纹在精神分裂(meta)与高血压(综述)中与疾病存在小但可重复的关联;皮纹遗传决定系数高、终身不变。
部分验证: 皮纹-疾病关联的跨研究结果不一致(有完全阴性报告),说明单用皮纹判别力不足;皮纹与 PRS 的信息重叠程度尚无定量估计。
待验证: 皮纹+PRS 联合模型的增量价值(命题一);皮纹异常在低 PRS 者中的独立预测力(命题二);皮纹-疾病关联的 PRS 中介比例(命题三)。
相关阅读
证据等级: B
参考文献: PMID 32273609, PMID 40063714, PMID 33900274, PMID 23116885, PMID 26265377, PMID 35313410
免责说明:本文属传统知识与科学文献的整理、对照与研究设计讨论,仅供理解与参考,不构成医疗建议,不能替代专业诊断与治疗。文中涉及的效应量、频率区间与暴露结论多为研究报告层面的推断,个体情况差异很大;如有身体不适或健康疑虑,请就医并由专业医师判断。
Dermatoglyphics and Genetic Risk Scores: Can Two \”Born with It\” Information Channels Jointly Predict Disease?
Evidence level: B
Genetic risk scores (PRS) are established to enhance disease prediction; dermatoglyphics represent another “fixed at birth” information channel. What is the incremental value of combining these two channels?
Main Text
First, observe an anomalous phenomenon. Dermatoglyphics (ridge structures in fingerprints and palm lines) predicting disease yields contradictory literature conclusions: a meta-analysis on schizophrenia shows patients have significantly lower total and a-b ridge counts than controls (small but reliable effect), while a systematic review on hypertension reports higher frequency of whorls and total ridge counts in patients. Conversely, one prospective study with 300 participants reported no statistical association between dermatoglyphics and type 2 diabetes or hypertension. Why does the same indicator sometimes work and sometimes not?
The answer may lie in a recent decade-long hot topic in predictive medicine: Polygenic Risk Scores (PRS). The essence of PRS is to weight-sum thousands of micro-effect risk loci to generate a score describing an individual’s “polygenic genetic burden.” Research from the Finnish FinnGen biobank (135,000 individuals) shows that high PRS carriers have a lifetime risk 21%–38% higher than those with average PRS, with onset occurring on average 4–9 years earlier, and significantly improved risk stratification in diabetes, atrial fibrillation, breast cancer, and prostate cancer. An analysis of 32 complex diseases in the UK Biobank provides more systematic figures: adding PRS to clinical prediction models improves average predictive accuracy by approximately 6%; heart failure sees the C-index jump from 0.546 to 0.751 (+37.5%), with stronger effects in early-onset populations. In short, the incremental value of PRS has been confirmed by large-scale biobanks, although this “increment” is mostly “small but real.”
PRS has a repeatedly emphasized selling point: quantifiable at birth, lifelong unchanged, and preceding all clinical risk factors. This mirrors dermatoglyphics almost exactly—dermatoglyphics become fixed during the embryonic period (weeks 10–24), remain unchanged throughout life, and have a heritability coefficient exceeding 0.80; they are also a “fixed at birth” individual marker. However, there is a key difference: PRS only captures the “genetic burden hard-coded in the DNA sequence,” whereas dermatoglyphics, as morphological imprints, additionally capture the interaction results between genes and intrauterine environments during embryogenesis. Hypoxia, nutrition, and hormonal fluctuations in early pregnancy can rewrite ridge development trajectories. In other words, dermatoglyphics may carry non-genetic but fixed-at-birth imprints that PRS cannot read.
This is the starting point of the joint hypothesis. If dermatoglyphics and PRS represent two partially independent information channels, the incremental value of a joint model depends on their correlation: complete correlation yields no increment, while zero correlation yields the maximum increment. Reality should lie between these extremes—dermatoglyphics share some developmental pathways with disease risk genes (which explains why dermatoglyphics GWAS loci enrich limb development pathways), but the portion of intrauterine environmental imprints is likely independent. The inconsistency in using dermatoglyphics alone to predict disease may stem from its weak signal being drowned out by small sample sizes and small effects; PRS provides a stable genetic backbone, while dermatoglyphics may supplement “environmental modification” branch information outside this backbone.
This leads to three testable propositions. Proposition 1: In cohorts for schizophrenia, hypertension, and coronary heart disease, is the ΔAUC and net reclassification improvement of a “dermatoglyphics+PRS” joint model relative to PRS alone significantly positive—this is a direct test of incremental value. Proposition 2: Do individuals with low PRS but abnormal dermatoglyphics (e.g., low a-b ridge counts, increased fluctuating asymmetry) still exhibit higher disease risk than those with low PRS and normal dermatoglyphics—testing whether dermatoglyphics carry intrauterine environmental imprints independent of genetic burden. Proposition 3: What proportion of the dermatoglyphics-disease association is mediated by PRS versus independent of PRS—using mediation analysis to separate the “genetic common factor” and “environmental imprint” pathways.
Evidence and Limitations
Verified: PRS provides real incremental value beyond clinical risk factors (5 conditions in FinnGen, 32 conditions in UK Biobank; Level A); dermatoglyphics show small but reproducible associations with disease in schizophrenia (meta-analysis) and hypertension (systematic review); dermatoglyphics have high heritability and are lifelong unchanged.
Partially Verified: Inconsistent cross-study results for dermatoglyphics-disease associations (including complete negative reports) indicate insufficient discriminative power when using dermatoglyphics alone; the degree of information overlap between dermatoglyphics and PRS lacks quantitative estimation.
To be Verified: Incremental value of the dermatoglyphics+PRS joint model (Proposition 1); independent predictive power of abnormal dermatoglyphics in low PRS individuals (Proposition 2); proportion of the dermatoglyphics-disease association mediated by PRS (Proposition 3).
Related Reading
- Cross-ethnic reproducibility of dermatoglyphic deep features and constitution clustering
- Quantitative mapping of dermatoglyphic features to modern phenotype clustering
- Evidence audit for dermatoglyphics and constitution associations
- Research pool on biofrequency codes
Evidence Level: B
References: PMID 32273609, PMID 40063714, PMID 33900274, PMID 23116885, PMID 2626377, PMID 35313410
Disclaimer: This article compiles traditional knowledge and scientific literature for comparison and research-design discussion only. It is not medical advice and cannot replace professional diagnosis or treatment. Effect sizes, frequency ranges and exposure conclusions cited here are largely inferences at the research-report level, and individual circumstances vary widely. If you have any discomfort or health concerns, please consult a physician.
Leave a Reply