中药材品质变异是物种对不同生态环境长期适应与自然选择的结果。遗传因素是中药材品质变异的生物学实质，环境因素是中药材品质变异的生态学实质。我国地理气候条件复杂多样，不同产地的同种药材品质差异显著，形成了特有的中药多道地、多产区现象。本文应用化学-遗传-生态相关性的综合分析新体系，运用转录组学技术、代谢组学技术、DNA条形码分子标记技术、UPLC-QTOF/MS技术并结合《药用植物全球产地生态适宜性信息系统》(Geographic Information System for Global Medicinal Plants，GMPGIS)，对肉苁蓉、黄花蒿品质变异机理进行综合分析。主要研究结论如下：

1. 对肉苁蓉进行品质变异分析，发现盐碱地、草地、沙地三种生境的肉苁蓉代谢物含量有显著差异，其中苯乙醇苷类成分总量呈现依次降低趋势。采用高通量测序技术对盐碱地、草地、沙地3种不同生境的肉苁蓉转录组测序，分别获得7.54G、7.50G、7.52G的原始数据。对转录本进行从头拼接获得52043条Unigenes，长度从201bp到15544bp不等，转录本长度N50为1362bp。对差异表达基因进行注释，找到苯丙氨酸解氨酶（PAL）、肉桂酸-4-羟化酶（C4H）、4-香豆酸-CoA连接酶（4CL）等11个苯乙醇苷类合成关键酶基因的候选基因。通过单核苷酸变异（SNP）的检测与注释，发现36条含SNP的Unigene注释到苯丙烷生物合成途径，其中5条为PAL候选基因，3条为4CL候选基因。这些基因的差异表达与变异可能是导致肉苁蓉品质差异形成的生物学原因。应用GMPGIS 技术对肉苁蓉进行适宜性区划和生态因子研究，发现中国、哈萨克斯坦、蒙古为适合肉苁蓉生长的潜在适宜区，年均日照、相对湿度对肉苁蓉生长影响较大。此外，对肉苁蓉的DNA条形码、转录组、代谢组与生态数据建立数据库，有利于数据的管理与共享。

2. 对黄花蒿进行品质变异分析，发现全国不同地区青蒿素含量有较大差异。秦岭-淮河分界线以南地区黄花蒿中青蒿素含量较高，并且形成以武陵山区为中心向四周扩散减低趋势。而分界线以北地区青蒿素含量普遍较低。运用DNA条形码分子标记技术对黄花蒿的遗传变异进行分析，发现黄花蒿种质资源存在着显著的遗传差异和丰富的多样性，且遗传多样性出现一定地域相关性。秦岭-淮河分界线以南地区遗传多样性有尤为丰富，这与青蒿素含量的分布趋势相似。利用GMPGIS技术平台预测高品质黄花蒿的潜在分布区，发现中国秦岭-淮河分界线以南地区为高品质黄花蒿最大的生态适宜区。光照、湿度等生态因子是影响黄花蒿品质的关键生态因子。该分析结果为更好地解释肉苁蓉、黄花蒿品质变异的形成机制提供了新思路，也为中药材品质变异理论研究提供科学依据。

The quality variation of Chinese traditional medicine is the result of long-term adaptation and natural selection of species to different ecological environment. Genetic factors are the biological essence of quality variation, and environmental factors are the ecological mechanism of the quality variation. The geographical and climatic conditions of our country are complex and diverse, and the quality of the same medicinal materials from different habitats is remarkable, forming a unique phenomenon of Chinese traditional medicine. In this paper, the quality variation mechanism of Cistanche deserticola and Artemisia annua was analyzed by the new system of comprehensive analysis of chemical-genetic-ecological correlation, integrating transcriptome, metabonomics, DNA barcoding, UPLC-QTOF/MS and Geographic Information System for Global Medicinal Plants(GMPGIS). The main results are as follows:

1. The primal data of 7.54G, 7.50G and 7.52G were sequenced by high throughput sequencing technology for 3 different habitats Cistanche deserticola of saline alkali, grassland and desert. 52043 transcripts were obtained from the transcript. The length of the transcript ranged from 201bp to 15544bp, and the N50 length of transcript was 1362bp. Candidate genes of 11 key enzyme genes including Phenylalanine ammonia-lyase (PAL), Cinnamate 4-hydroxylase (C4H) and Cinnamate4-hydroxylas (4CL) were found.  Through the detection and annotation of single nucleotide variation (SNP), 36 SNP containing Unigenes annotated to phenylpropane biosynthesis pathway were found, of which 5 unigenes were PAL candidate genes and 3 unigenes were 4CL candidate genes. The differential expression and variation of these genes may be the biological reasons leading to the difference in quality of Cistanche deserticola. The suitability regionalization and ecological factors of Cistanche deserticola were studied by GMPGIS technology. It was found that China, Kazakhstan and Mongolia are potential suitable areas for the growth of Cistanche deserticola. The annual average sunshine and the relative humidity had great influence on the growth of Cistanche deserticola. In addition, the establishment of the database for DNA barcoding, transcriptome, metabonomics and ecological data of Cistanche deserticola data is conducive to data management and sharing.

2. The analysis of quality variation of Artemisia annua showed that there were great differences in artemisinin content in different parts of the country. Artemisinin content of Artemisia annua in the south of Qinling Mountains Huaihe demarcation line is relatively high, and the trend of spreading around Wuling mountain area is decreasing. However, the content of artemisinin in the north of the boundary is generally low. The genetic variation of Artemisia annua was analyzed by DNA barcoding. It was found that there were significant genetic differences and rich diversity in the germplasm resources of Artemisia annua, and the genetic diversity of Artemisia annua was related to a certain region. The genetic diversity in the south of Qinling Mountains Huaihe demarcation line is particularly rich, which is similar to the distribution trend of artemisinin content. The potential distribution area of high quality Artemisia annua was predicted by GMPGIS. It was found that the south area of Qinling Mountains Huaihe demarcation line was the largest ecological suitable area for the high quality Artemisia annua. Sunshine and the relative humidity are the key ecological factors affecting the quality of Artemisia annua. The analysis results provide a new idea to better explain the formation mechanism of the quality variation of Cistanche deserticola and Artemisia annua, and provide scientific basis for the study of the theory of quality variation of Chinese traditional medicine.