Evidence level: B
上一篇用流行病学的效应量说:出生季节的影响随日照对比度(纬度)单调变化。但末端的疾病指标最容易被混杂污染。要证明这是真规律,得看分子层:同一批季节相关的甲基化位点,在高纬和低纬队列里的差异方向和幅度,是不是也照着同一条曲线走。
正文
上一步留下的软肋
前一篇把各疾病出生季效应量统一回归到”冬夏日照对比度”这个连续变量上,得到可证伪命题:效应量随对比度单调递增。多发性硬化的挪威南北梯度、精神分裂症 27 个地点的纬度相关都支持这条曲线。
但软肋在于:用的全是”疾病发生率”这类末端指标。诊断习惯、就医时机、乃至出生登记本身的季节波动,都能伪造出一条漂亮的纬度梯度。要堵这个洞,只能下沉一层——看分子层的印记,是否也在同一方向上随纬度变化。
为什么分子层是独立证据
DNA 甲基化正好合适。它稳定(胚胎早期确立后可在血里读到多年)、系统(亚稳态表观等位基因全身组织保持同一状态,抽一管血可推断内脏印记,PMID 21203497)、且有已知的响应位点(可预先注册、定向检验,而非全基因组漫游)。于是问题变硬了:
同一批”出生季节敏感”的甲基化位点,在高纬与低纬队列中,甲基化差异的符号与幅度是否随日照对比度单调变化?
三条证据线
第一条:跨队列 meta 已经看到”纬度效应”。 2023 年 PACE 联盟合并 21 个出生队列(N = 9,358,脐带血)与 12 个儿童队列(N = 3,610,1–11 岁)的甲基化数据,做出生季节全表观组 meta,再按纬度分层(≥50°N 与 <50°N)重跑一遍(PMID 37697338)。结果出现两种效应:时间效应(出生时 2 个 CpG / 26 个差异甲基化区域 DMR;儿童期 4 个 CpG / 32 个 DMR)与空间效应(分层后关联模式改变,若干 DMR 只在 ≥50°N 出现,指向已报道与精神分裂症(LAX1)、皮肤疾病(PSORS1C)、气道炎症与哮喘(LTB4R)相关的基因)。这是第一次把”出生季节 × 甲基化”和纬度放进同一模型。但同篇自曝:不同季节之间、出生期与儿童期之间 DMR 重叠极少——位点集本身还不稳定。
第二条:低纬度不是没有印记,而是换了驱动源。 冈比亚(约 13°N)全年光照几乎不波动,但有剧烈的雨季/旱季营养与感染季节。雨季受孕者若干亚稳态表观等位基因甲基化系统性升高,且稳定至少十年(PMID 21203497);全基因组筛选进一步锁定 VTRNA2-1(nc886)为最稳健的环境响应表观等位基因(PMID 26062908),其印记频率与受孕季节相关,并关联到 5 岁时的 BMI(PMID 30509985);过敏领域也观察到出生季节与甲基化的同向关联(PMID 26973132)。含义是:低纬印记的驱动变量从日照换成了营养/感染,方向可能反而与高纬光照印记相反——这正是”机制切换”最直接的可测表现。
第三条:必须防的细胞组成陷阱。 外周血甲基化是多种白细胞的混合读数,而白细胞比例本身就有季节波动。有研究用同一批人的全血、颊上皮、鼻上皮加荧光分选的五种白细胞做交叉比对,证实细胞类型差异会污染全血信号(PMID 33739972)。所以,一个”看起来随季节变化”的位点,可能只是白细胞谱的季节漂移。未做细胞组成校正的季节甲基化信号,不能算分子层证据。
一个自带证伪条件的检验设计
综合三条线,能一次性区分”真印记/机制切换/技术假象”的设计是清楚的:①统一协议(脐带血、同芯片平台、同细胞组成校正管道);②预先注册候选位点集(VTRNA2-1/nc886、CACNA1C、LAX1、PSORS1C、LTB4R);③两个结局——方向一致性(各队列 Δβ 符号是否守恒)与幅度(|Δβ| 对日照对比度回归);④三种结果三种解释:幅度单调上升且方向守恒 → 支持统一剂量-反应模型;低纬方向反转 → 支持机制切换;幅度与纬度无关、却与采血月份或某类白细胞比例相关 → 判为技术假象。
这套设计的最大好处是自带证伪条件:若分子层的方向一致性完全守不住,上一篇文章那条漂亮的纬度梯度就得退回去重新怀疑混杂。
一句话收尾:易学讲”时令旺衰决定禀赋”,说的是一套环境写入命运的机制;甲基化恰好是这套机制在分子层的”笔迹”。笔迹是否随纬度换手感、换方向,比任何一句”古书这么说”更能说明它到底是自然规律,还是文化叙事。
证据与局限
已验证: 出生季节与 DNA 甲基化存在全表观组关联,按纬度分层后出现空间效应,若干 DMR 只在 ≥50°N 高纬队列出现(PMID 37697338);低纬(冈比亚)受孕季节通过营养/感染通路系统性改变亚稳态表观等位基因甲基化,稳定 ≥10 年(PMID 21203497);VTRNA2-1(nc886)被独立筛选确认为最稳健的环境响应表观等位基因(PMID 26062908),其印记频率与受孕季节相关并关联儿童 BMI(PMID 30509985);出生季节与甲基化、过敏病存在同向关联(PMID 26973132);细胞类型组成会显著影响全血甲基化信号(PMID 33739972)。
部分验证: “同一批位点跨纬度方向一致”尚未被直接检验——meta 自报不同季节、不同时点间 DMR 重叠极少(PMID 37697338),位点集尚不稳定;低纬印记由营养还是感染主导、方向是否与高纬光照印记相反,目前只有间接提示。
待验证: ①统一协议与细胞组成校正下,Δβ 符号在高、低纬队列间是否守恒;②若低纬方向反转,拐点纬度(机制切换阈值)落在哪里;③被判定”季节敏感”的位点,有多少能在细胞组成校正后存活。
相关阅读
证据等级: B
参考文献: PMID 37697338, PMID 21203497, PMID 26062908, PMID 30509985, PMID 26973132, PMID 33739972
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。
Do \”methylation notes\” written in the genes of birth season change direction with latitude?
Evidence level: B
The previous article used epidemiological effect sizes to argue that birth-season effects vary monotonically with solar contrast (latitude). However, endpoint disease metrics are most susceptible to confounding. To verify this as a genuine pattern, we must examine the molecular layer: Do the same season-sensitive methylation sites show differences in direction and magnitude that follow the same curve across high-latitude and low-latitude cohorts?
Main Text
The Weakness Left by the Previous Step
The previous article regressed disease birth-season effect sizes across all conditions onto a continuous variable: “winter-summer solar contrast,” yielding the falsifiable proposition: effect sizes increase monotonically with contrast. The Norwegian north-south gradient for multiple sclerosis and the latitude correlation across 27 sites for schizophrenia support this curve.
However, the weakness lies in the use of endpoint metrics like disease incidence. Diagnostic habits, timing of medical visits, and even seasonal fluctuations in birth registration can fabricate a convincing latitude gradient. To plug this hole, we must descend one layer—examining whether molecular markers also shift in the same direction with latitude.
Why the Molecular Layer Provides Independent Evidence
DNA methylation is precisely suited for this. It is stable (established early in embryogenesis and readable in blood for years), systemic (metastable epialleles maintain the same state across tissues; drawing one blood tube can infer internal organ marks, PMID 21203497), and has known response loci (pre-registered and targeted testing is possible, rather than genome-wide roaming). Thus, the question becomes sharper:
Do the same set of “birth-season sensitive” methylation sites show methylation differences that vary monotonically with solar contrast in terms of both sign and magnitude across high-latitude and low-latitude cohorts?
Three Lines of Evidence
Line 1: Cross-cohort meta-analysis already reveals a “latitude effect.” In 2023, the PACE consortium merged methylation data from 21 birth cohorts (N = 9,358, cord blood) and 12 pediatric cohorts (N = 3,610, ages 1–11) for a birth-season-wide epigenome meta-analysis, followed by re-running the analysis stratified by latitude (≥50°N vs. <50°N) (PMID 37697338). Two effects emerged: temporal effects (2 CpGs / 26 DMRs at birth; 4 CpGs / 32 DMRs in childhood) and spatial effects (altered association patterns after stratification; several DMRs appeared only in ≥50°N, pointing to genes previously linked to schizophrenia (LAX1), skin disease (PSORS1C), airway inflammation and asthma (LTB4R)). This is the first model to combine “birth season × methylation” with latitude. However, the same paper self-admitted: DMR overlap is minimal between different seasons and between birth and childhood—the site set itself remains unstable.
Line 2: Low latitudes are not devoid of marks, but driven by different sources. In Gambia (~13°N), sunlight fluctuates little year-round, but there are intense rainy/dry season nutritional and infection cycles. Pregnant individuals during the rainy season show systematic increases in several metastable epialleles, stable for at least ten years (PMID 21203497). Genome-wide screening further identified VTRNA2-1 (nc886) as the most robust environmental response epiallele (PMID 26062908); its imprinting frequency correlates with conception season and associates with BMI at age 5 (PMID 30509985). Similar co-directional associations between birth season and methylation were observed in allergy research (PMID 26973132). The implication is: the driving variable for low-latitude marks shifts from sunlight to nutrition/infection, potentially reversing the direction relative to high-latitude light marks—this is the most direct measurable manifestation of a “mechanism switch.”
Line 3: The cellular composition trap to avoid. Peripheral blood methylation is a mixed readout of multiple leukocyte types, and leukocyte proportions themselves fluctuate seasonally. Studies comparing whole blood, buccal epithelium, nasal epithelium, and five fluorescently sorted leukocyte types from the same individuals confirmed that cell-type differences contaminate whole blood signals (PMID 33739972). Therefore, a site that “appears to vary by season” may simply reflect seasonal drift in the leukocyte profile. Seasonal methylation signals without cellular composition correction cannot be considered molecular layer evidence.
A Validation Design with Built-in Falsifiability
Synthesizing the three lines, a design capable of distinguishing “true marks/mechanism switch/technical artifact” is clear: ① Unified protocol (cord blood, same chip platform, same cell composition correction pipeline); ② Pre-registered candidate site set (VTRNA2-1/nc886, CACNA1C, LAX1, PSORS1C, LTB4R); ③ Two outcomes—directional consistency (whether Δβ signs are conserved across cohorts) and magnitude (regression of |Δβ| against solar contrast); ④ Three results, three interpretations: monotonic increase in magnitude with conserved direction → supports a unified dose-response model; reversal of direction at low latitudes → supports mechanism switch; magnitude unrelated to latitude but correlated with blood draw month or specific leukocyte proportions → classified as technical artifact.
The greatest advantage of this design is that it carries its own falsification condition: if molecular layer directional consistency cannot be maintained, the elegant latitude gradient from the previous article must be retracted and re-evaluated for confounding.
In summary: Yijing Studies posits that “seasonal prosperity determines constitution,” describing a mechanism of environmental inscription into fate. Methylation is precisely the “handwriting” of this mechanism at the molecular level. Whether the handwriting changes its feel or direction with latitude speaks more clearly than any “ancient text says” whether it is a natural law or a cultural narrative.
Evidence and Limitations
Verified: Birth season associates with DNA methylation genome-wide; spatial effects emerge after latitude stratification, with several DMRs appearing only in ≥50°N high-latitude cohorts (PMID 37697338); in low latitudes (Gambia), conception season systematically alters metastable epiallele methylation via nutritional/infection pathways, stable ≥10 years (PMID 21203497); VTRNA2-1 (nc886) independently confirmed as the most robust environmental response epiallele (PMID 26062908), with imprinting frequency correlating with conception season and associating with childhood BMI (PMID 30509985); birth season associates directionally with methylation and allergy (PMID 26973132); cell type composition significantly affects whole blood methylation signals (PMID 33739972).
Partially Verified: “Consistent direction across latitudes for the same set of sites” has not been directly tested—the meta-analysis reports minimal DMR overlap between different seasons and time points (PMID 37697338), indicating site set instability; whether low-latitude marks are driven by nutrition or infection, and whether the direction opposes high-latitude light marks, remains only indirectly suggested.
To Be Verified: ① Whether Δβ signs are conserved between high- and low-latitude cohorts under unified protocols and cell composition correction; ② If a reversal occurs at low latitudes, where the inflection latitude (mechanism switch threshold) lies; ③ How many sites deemed “season-sensitive” survive after cellular composition correction.
Related Reading
- Dose-response relationship between birth season effects and solar contrast
- Patterns of variation in birth season effects across latitude gradients
- Nayin Five Elements and Seasonal Phases of Gut Microbiota
- Research Pool 1: Environment-Life Frequency Interface
Evidence Level: B
References: PMID 37697338, PMID 21203497, PMID 26062908, PMID 30509985, PMID 26973132, PMID 33739972
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