Evidence level: B
多基因风险评分(PRS)只读了”基因代码”,指纹却在胚胎第 4-20 周就永久定格了”发育环境日志”——低 PRS 人群中的皮纹异常者,可能正是被基因评分漏掉的高危个体。
正文
手相文化里有一句流传千年的直觉:指纹藏着一个人一生的秘密。现代科学给了这句直觉一个意外的支点——指纹纹脊的形成窗口(胚胎 4-20 周)恰好与器官发生、代谢编程的关键期重叠,且一经刻下便终生不变。于是出现一个现代遗传学难以回避的盲区:当我们只用 DNA 变异算出的多基因风险评分(PRS)给人划分风险等级时,那些发育早期环境扰动留下的、写在指纹上的痕迹,会不会恰好是 PRS 这个”基因代码阅读器”看不见的部分?
先看清 PRS 的能力边界。2023 年 BMJ 发表的 Polygenic Score Catalog 二次分析(PMID 37859783)给 PRS 泼了盆冷水:在人群筛查、个体预测、风险分层三个场景下,PRS 表现均有限——即便在 PRS 最高 2.5% 的人群里,10 年冠心病发病比仍只有约 1:7,而最低五分位组的人群 10 年比也高达 1:41,仍有人发病。换句话说,处于”低遗传风险”分层的个体,并非真的安全,疾病仍会在那里出现。这指向一个结构性问题:常见病的遗传度远未被 GWAS 位点解释,大量”遗传之外的风险”无处安放。
PRS 漏掉的部分到底装在哪?FinnGen 30.6 万人的系统比较(PMID 36347255)给出了关键的定量答案:第一级亲属家系史与 PRS 基本独立、互为补充,PRS 平均只解释了家系史效应的 10%,而家系史只解释了 PRS 效应的 3%。既然家系史里约九成的风险信息是 PRS 读不到的,那么这些被漏掉的信息显然包含大量”非单核苷酸变异”的成分——共享环境、基因-环境交互,以及早期发育编程。而指纹,恰好是这个”发育编程”留下的物理痕迹。
这正是荷兰饥荒研究的价值所在。Kahn 等在 1944-45 年荷兰冬季饥荒队列中发现,母亲妊娠早期的营养环境会永久改变子女的指纹特征,且独立于出生体重(PMID 17929242)。更重要的是,同样的指纹标志(拇指与尾指脊线数差 Md15)与中老年糖尿病独立相关,校正年龄、性别、父母糖尿病与成人体型后 OR=1.37(PMID 18684786)——一个孕早期定型的指标,竟能预测 50 年后发病。这条证据链打通了”宫内环境 → 指纹定型 → 成年疾病”的完整路径,也暗示:皮纹携带的风险信号,很可能有一部分与常染色体 SNP 无关,属于”宫内印记”通道。
把两条证据链接起来,就得到可检验的核心命题:在 PRS 处于低分位的个体中,皮纹异常者是否仍携带独立的疾病风险? 若成立,则皮纹提供了 PRS 与家系史之外的第三类信息——不是”遗传负荷的另一个代理”,而是”宫内发育环境的直接化石”。这与中医”先天之精”的直觉惊人地同构:指纹是先天禀赋的可测量表达,而先天不仅由基因决定,还由孕期的环境共同雕刻。
设计上,最干净的检验是 2×2 分层:按 PRS 中位切分”低/高遗传负荷”,再按皮纹定量特征(斗纹频率、脊线总数 TABRC、Md15、ATD 角)切分”皮纹正常/异常”,在精神分裂、高血压、冠心病、2 型糖尿病等多病种队列中建模,核心看交互项与低 PRS×皮纹异常亚组的独立 HR。判定标准应预设:若低 PRS 人群中皮纹异常者风险显著高于皮纹正常者(相对危险比 ≥1.3 且置信区间不含 1),并经多病种重复与族裔分层稳定,则”皮纹携带独立于遗传负荷的宫内印记风险”成立。
需要警惕的两类替代解释:一是剩余遗传混杂——低 PRS 组内仍可能存在未被 GWAS 捕获的罕见变异,皮纹异常只是这些变异的表面相关;二是测量混杂——族裔基线差异可能让”皮纹异常”与”高发病”的关联是伪相关。因此方案必须内置族裔分层、PRS 训练集外部验证、以及同胞对(同一宫内环境、共享一半基因)的 discordant 分析来剥离环境与遗传。
证据与局限
已验证:
– PRS 在人群层面风险分层能力有限,低分位组仍有可观发病率(PMID 37859783)
– 家系史与 PRS 独立互补,PRS 仅解释家系史效应约 10%(PMID 36347255)
– 宫内早期营养环境永久改变指纹特征(PMID 17929242)
– 指纹标志 Md15 独立于出生体重预测中年糖尿病(PMID 18684786)
部分验证:
– “低 PRS + 皮纹异常 → 独立疾病风险”尚无直接队列证据,为待检验假说
– 皮纹信号有多少落在遗传负荷之外,缺乏定量拆分
待验证:
– 低 PRS×皮纹异常交互项在精神分裂/高血压/冠心病/2 型糖尿病中的方向与幅度
– 同胞对 discordant 分析能否剥离宫内环境与遗传共因
– 族裔分层后皮纹-风险关联的残余稳健性
相关阅读
证据等级: B(PRS 局限性与指纹-宫内印记均有直接实证,但”低 PRS×皮纹异常独立风险”为间接证据支撑的实验设计方案)
参考文献: PMID 37859783, PMID 36347255, PMID 35460399, PMID 18684786, PMID 17929242, PMID 9637128
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。
Low Polygenic Risk Score but Abnormal Dermatoglyphics: Can Fingerprints Reveal Intrauterine Markets Missed by Genetic Scores Beyond Genetic Load?
Evidence level: B
Polygenic Risk Scores (PRS) read only the “genetic code,” yet fingerprints permanently encode a “developmental environment log” during embryonic weeks 4–20. Among individuals with low PRS, those with abnormal dermatoglyphics may represent high-risk cases overlooked by gene scoring.
Main Text
Within palmistry culture, a perennial intuition holds that fingerprints conceal the secrets of a lifetime. Modern science provides an unexpected pivot for this intuition: the formation window for fingerprint ridges (embryonic weeks 4–20) overlaps precisely with critical periods of organogenesis and metabolic programming, and once imprinted, remains unchanged for life. This creates a blind spot difficult to avoid in modern genetics: when we assign risk levels based solely on Polygenic Risk Scores (PRS) derived from DNA variants, will the traces left by early developmental environmental perturbations—inscribed on fingerprints—represent a portion invisible to PRS, the “genetic code reader”?
First, clarify the capability boundaries of PRS. A secondary analysis of the Polygenic Score Catalog published in BMJ in 2023 (PMID 37859783) cooled expectations across three scenarios: population screening, individual prediction, and risk stratification. Even within the top 2.5% of PRS distribution, the 10-year incidence ratio for coronary heart disease remains approximately 1:7, while the lowest quintile group shows a ratio of 1:41, yet individuals still develop the disease. In other words, individuals categorized as having “low genetic risk” are not truly safe; disease still manifests there. This points to a structural issue: the heritability of common diseases is far not explained by GWAS loci, leaving vast amounts of “risk beyond genetics” unaccounted for.
Where exactly is the missing portion stored? A systematic comparison of 306,000 individuals in FinnGen (PMID 36347255) provided a key quantitative answer: family history in first-degree relatives is largely independent of and complementary to PRS. PRS explains only about 10% of the family history effect, whereas family history explains only 3% of the PRS effect. Since approximately 90% of risk information in family history remains unread by PRS, this omitted information clearly contains substantial components of “non-single-nucleotide variants”—shared environments, gene-environment interactions, and early developmental programming. Fingerprints, coincidentally, are the physical traces left by this “developmental programming.”
This is precisely where the Dutch Famine Study holds value. Kahn et al. found in their 1944–1945 Dutch winter famine cohort that maternal nutritional environments during early pregnancy permanently alter offspring fingerprint characteristics, independent of birth weight (PMID 17929242). More importantly, the same fingerprint marker (difference in ridge count between thumb and little finger, Md15) shows an independent association with adult-onset diabetes in midlife. After adjusting for age, sex, parental diabetes, and adult body size, the odds ratio (OR) is 1.37 (PMID 18684786)—an indicator定型 in early pregnancy predicting onset 50 years later. This evidence chain connects the complete pathway from “intrauterine environment → fingerprint fixation → adult disease,” also suggesting that risk signals carried by dermatoglyphics likely belong to a channel unrelated to autosomal SNPs, termed an “intrauterine imprint.”
Linking these two evidence streams yields a testable core proposition: Among individuals in the low PRS percentile, do those with abnormal dermatoglyphics still carry independent disease risk? If true, fingerprints provide a third category of information beyond PRS and family history—not merely “another proxy for genetic load,” but a “direct fossil of intrauterine developmental environment.” This resonates strikingly with the TCM intuition of “Constitutional Essence” (Xian Tian Zhi Jing): fingerprints are a measurable expression of innate endowment, while the innate is shaped not only by genes but also co-carved by prenatal environments.
In design, the cleanest test is a 2×2 stratification: split by median PRS into “low/high genetic load,” then split by quantitative dermatoglyphic features (whorl frequency, total ridge count TABRC, Md15, ATD angle) into “normal/abnormal dermatoglyphics.” Model across multicohort studies for multiple conditions (schizophrenia, hypertension, coronary heart disease, type 2 diabetes), focusing on the interaction term and independent Hazard Ratios (HR) in the “low PRS × abnormal dermatoglyphics” subgroup. Pre-set criteria should include: if risk in low PRS individuals with abnormal dermatoglyphics is significantly higher than those with normal dermatoglyphics (relative risk ratio ≥1.3 with confidence intervals excluding 1), and this finding is replicated across multiple diseases and stratified by ethnicity, then “intrauterine imprint risk carried by dermatoglyphics independent of genetic load” is established.
Two types of alternative explanations require vigilance: first, residual genetic confounding—rare variants not captured by GWAS may still exist within the low PRS group, with abnormal dermatoglyphics merely being a surface correlation; second, measurement confounding—ethnic baseline differences might render the association between “abnormal dermatoglyphics” and “high incidence” spurious. Therefore, the protocol must embed ethnic stratification, external validation of PRS training sets, and discordant analysis of twin pairs (sharing intrauterine environments and half their genes) to disentangle environmental and genetic factors.
Evidence and Limitations
Verified:
– PRS demonstrates limited risk stratification capability at the population level, with significant incidence rates even in low-percentile groups (PMID 37859783)
– Family history and PRS are independent and complementary; PRS explains only ~10% of the family history effect (PMID 36347255)
– Early intrauterine nutritional environments permanently alter fingerprint characteristics (PMID 17929242)
– The fingerprint marker Md15 predicts midlife diabetes independently of birth weight (PMID 18684786)
Partially Verified:
– No direct cohort evidence yet exists for “Low PRS + Abnormal Dermatoglyphics → Independent Disease Risk”; this remains a hypothesis awaiting testing
– The proportion of dermatoglyphic signals falling outside genetic load lacks quantitative decomposition
Pending Verification:
– Direction and magnitude of the low PRS × abnormal dermatoglyphics interaction term in schizophrenia, hypertension, coronary heart disease, and type 2 diabetes
– Whether discordant twin pair analysis can disentangle intrauterine environments from shared genetic factors
– Residual robustness of dermatoglyphics-risk associations after ethnic stratification
Related Reading
- Joint Model of Dermatoglyphics + PRS vs. PRS Alone for Incremental Disease Prediction
- Incremental Predictive Value of Dermatoglyphics + PRS
- Mendelian Causality of TCM Constitution PRS
- Research Pool: Biological Frequency Code
Evidence Level: B (PRS limitations and fingerprint-intrauterine imprint have direct empirical support, but “low PRS × abnormal dermatoglyphics independent risk” is supported by an indirect experimental design proposal)
References: PMID 37859783, PMID 36347255, PMID 35460399, PMID 18684786, PMID 17929242, PMID 9637128
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