School of Life Sciences, Northwestern Polytechnical University, Xi'an, 70072, China.
School of Software and Microelectronics, Northwestern Polytechnical University, Xi'an, 70072, China.
BMC Bioinformatics. 2018 Aug 13;19(Suppl 9):281. doi: 10.1186/s12859-018-2274-3.
Human Microbiome Project reveals the significant mutualistic influence between human body and microbes living in it. Such an influence lead to an interesting phenomenon that many noninfectious diseases are closely associated with diverse microbes. However, the identification of microbe-noninfectious disease associations (MDAs) is still a challenging task, because of both the high cost and the limitation of microbe cultivation. Thus, there is a need to develop fast approaches to screen potential MDAs. The growing number of validated MDAs enables us to meet the demand in a new insight. Computational approaches, especially machine learning, are promising to predict MDA candidates rapidly among a large number of microbe-disease pairs with the advantage of no limitation on microbe cultivation. Nevertheless, a few computational efforts at predicting MDAs are made so far.
In this paper, grouping a set of MDAs into a binary MDA matrix, we propose a novel predictive approach (BMCMDA) based on Binary Matrix Completion to predict potential MDAs. The proposed BMCMDA assumes that the incomplete observed MDA matrix is the summation of a latent parameterizing matrix and a noising matrix. It also assumes that the independently occurring subscripts of observed entries in the MDA matrix follows a binomial model. Adopting a standard mean-zero Gaussian distribution for the nosing matrix, we model the relationship between the parameterizing matrix and the MDA matrix under the observed microbe-disease pairs as a probit regression. With the recovered parameterizing matrix, BMCMDA deduces how likely a microbe would be associated with a particular disease. In the experiment under leave-one-out cross-validation, it exhibits the inspiring performance (AUC = 0.906, AUPR =0.526) and demonstrates its superiority by ~ 7% and ~ 5% improvements in terms of AUC and AUPR respectively in the comparison with the pioneering approach KATZHMDA.
Our BMCMDA provides an effective approach for predicting MDAs and can be also extended to other similar predicting tasks of binary relationship (e.g. protein-protein interaction, drug-target interaction).
人类微生物组计划揭示了人体与生活在其中的微生物之间存在着显著的互利共生影响。这种相互影响导致了一个有趣的现象，即许多非传染性疾病与各种微生物密切相关。然而，由于微生物培养的高成本和局限性，识别微生物-非传染性疾病关联（MDAs）仍然是一项具有挑战性的任务。因此，需要开发快速方法来筛选潜在的 MDAs。越来越多的已验证的 MDAs 使我们能够以新的视角满足需求。计算方法，特别是机器学习，有望在大量微生物-疾病对中快速预测 MDA 候选物，并且不受微生物培养的限制。然而，到目前为止，只有少数计算工作用于预测 MDAs。
在本文中，我们将一组 MDAs 分组到一个二进制 MDA 矩阵中，提出了一种基于二进制矩阵补全的新预测方法（BMCMDA）来预测潜在的 MDAs。所提出的 BMCMDA 假设，未观测到的 MDA 矩阵是潜在参数矩阵和噪声矩阵的总和。它还假设 MDA 矩阵中观测到的项的独立出现的下标遵循二项式模型。对于噪声矩阵，采用标准均值为零的高斯分布，我们将参数矩阵与观测到的微生物-疾病对之间的关系建模为概率回归。利用恢复的参数矩阵，BMCMDA 推断出一种微生物与特定疾病相关的可能性。在留一交叉验证实验中，它表现出令人鼓舞的性能（AUC=0.906，AUPR=0.526），与开创性的 KATZHMDA 方法相比，在 AUC 和 AUPR 方面分别提高了约 7%和 5%，证明了其优越性。
我们的 BMCMDA 为预测 MDAs 提供了一种有效的方法，并且还可以扩展到其他类似的二进制关系预测任务（例如，蛋白质-蛋白质相互作用，药物-靶标相互作用）。
