利用資料探勘探討大豆成分與乳癌之關係
No Thumbnail Available
Date
2018
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
乳癌是威脅全球女性健康甚至生命的疾病之一，流行病學研究顯示台灣乳癌發生率在近二十年有上升的趨勢。飲食是影響乳癌發展的因子之一，過去流行病學研究指出攝取大豆與降低乳癌風險的關係，但在部份體內與體外試驗中卻看到相反的結果。因此攝取大豆與降低乳癌風險的關係仍存有些爭議。大豆中有許多具生物活性的成份，其中包括植物性雌激素和具活性的蛋白、胜肽。而大豆異黃酮是大豆中最主要的植物性雌激素類成份，由於其結構與雌二醇相似，因此在不同條件下，可以扮演協同或是拮抗雌激素的作用。近年來發表了許多以電腦與資訊相關方法進行的生命科學研究，有助於精準醫學的發展。本研究利用資料探勘相關模型探討大豆攝取與降低乳癌風險之關係，使用的資料探勘模型包括決策樹分類模型和關聯規則探勘，建立一大豆與乳癌為主題之資料庫，其中包括了201篇原著研究及478筆資料。根據研究性質，將資料庫分成體內試驗、體外試驗與人類研究三個子資料庫，探討大豆中不同類別成分、其乳癌之類型以及激素存在與否不同條件下，大豆對乳癌發展的影響是否有所差異。以大豆成分類別來看，蛋白胜肽類成份對乳癌正面影響關聯性較高，以正面影響為主；而植物性雌激素類則有較多反向結果，其中以Daidzein對乳癌具有負面影響，其他大豆中植物雌激素類成分如Coumestrol亦對乳癌發展具有部分負面影響。此外分析結果也顯示大豆影響乳癌會因不同乳癌類型而有所差異，在Luminal類型的乳癌中看到大豆之影響多有不一致的結果，而三陰性乳癌則多有正面的影響。而針對不同受體表現分析結果顯示，大豆對ER+和PR+表現之乳癌會有不一致影響結果，其中是以植物性雌激素類成分有負面影響。而對ER-和PR-類型乳癌以正面影響為主。而在雌激素存在之下，如停經前婦女，大豆成分對乳癌的正面影響也有較強的關聯。總結來說，大豆影響乳癌之結果取決於多種不同的條件。因此在建議乳癌病人攝取大豆與其相關製品時，需要考慮其乳癌類型以及選擇特定之大豆成份。
Breast cancer is one of the diseases that threatens the health and even life of women around the world. Epidemiological studies have shown that the incidence of breast cancer has risen in the past 20 years in Taiwan. Dietary pattern is one of the factors to influence breast cancer development. Soy intake is associated to reduce the risk of breast cancer in some epidemiologic studies. However, there were some conflicting results in both in vitro and in vivo studies. It is still controversial between soy intake and the risk of breast cancer. There are abundant bioactive compounds in soy, including the phytoestrogen and bioactive proteins and peptides. The main phytoestrogen of soy are isoflavones, which have similar structure with 17-β-estradiol, and exerts estrogenic or antiestrogenic effects. Depend on different condition, isoflavones can effect as estrogen antagonist or agonist. Recently, scientists apply the computational and information-related methods to assist the development of precision medicine on life sciences. In this study, we use the data mining model including decision tree classification and association rule mining to analyize the relation between soy consumption and the risk of breast cancer. The database of soy-breast cancer containing 201 original research papers and 478 data, whcih was separated to three sub-databases included cell, animal and human studies. The aim of this study is to explore whether the effects of soy on breast cancer development are different under different conditions of different category compounds in soy, the type of breast cancer, and the presence or absence of hormones receptors. The results have shown that the category of protein or peptides is related to the positive effect on the breast cancer. However, the category of phytoestrogen has more conflicting results. Especially, we found that the Daidzein is the specific compound that related to negative effect to the breast cancer. Other phytoestrogen compounds such as Coumestrol also has some negative impact on breast cancer development. Soy has more conflict results on the breast cancer of luminal subtype but has more positive effects on the triple negative breast cancer. According to the results of different receptor performance analysis, soy has a negative effect on breast cancer of ER+ and PR+, which is especially phytoestrogen compounds have negative effect. And it is mainly positive for ER- and PR- type breast cancer. Regard to the presence of estrogen, soy has a stronger positive correlation with on premenopausal breast cancer women.with In conclusion, the association between soy and breast cancer is depended on variety conditions. Thus, the recommendation for breast cancer patients to take soy products should consider the tumor types and specific soy compounds.
Breast cancer is one of the diseases that threatens the health and even life of women around the world. Epidemiological studies have shown that the incidence of breast cancer has risen in the past 20 years in Taiwan. Dietary pattern is one of the factors to influence breast cancer development. Soy intake is associated to reduce the risk of breast cancer in some epidemiologic studies. However, there were some conflicting results in both in vitro and in vivo studies. It is still controversial between soy intake and the risk of breast cancer. There are abundant bioactive compounds in soy, including the phytoestrogen and bioactive proteins and peptides. The main phytoestrogen of soy are isoflavones, which have similar structure with 17-β-estradiol, and exerts estrogenic or antiestrogenic effects. Depend on different condition, isoflavones can effect as estrogen antagonist or agonist. Recently, scientists apply the computational and information-related methods to assist the development of precision medicine on life sciences. In this study, we use the data mining model including decision tree classification and association rule mining to analyize the relation between soy consumption and the risk of breast cancer. The database of soy-breast cancer containing 201 original research papers and 478 data, whcih was separated to three sub-databases included cell, animal and human studies. The aim of this study is to explore whether the effects of soy on breast cancer development are different under different conditions of different category compounds in soy, the type of breast cancer, and the presence or absence of hormones receptors. The results have shown that the category of protein or peptides is related to the positive effect on the breast cancer. However, the category of phytoestrogen has more conflicting results. Especially, we found that the Daidzein is the specific compound that related to negative effect to the breast cancer. Other phytoestrogen compounds such as Coumestrol also has some negative impact on breast cancer development. Soy has more conflict results on the breast cancer of luminal subtype but has more positive effects on the triple negative breast cancer. According to the results of different receptor performance analysis, soy has a negative effect on breast cancer of ER+ and PR+, which is especially phytoestrogen compounds have negative effect. And it is mainly positive for ER- and PR- type breast cancer. Regard to the presence of estrogen, soy has a stronger positive correlation with on premenopausal breast cancer women.with In conclusion, the association between soy and breast cancer is depended on variety conditions. Thus, the recommendation for breast cancer patients to take soy products should consider the tumor types and specific soy compounds.
Description
Keywords
大豆, 乳癌, 雌激素, 植物性雌激素, 決策樹, 關聯規則探勘, Soy, Breast cancer, Estrogen, Phytoestrogen, Decision tree, Association rule mining