QSAR models based on isomorphic and nonisomorphic data fusion for predicting the blood brain barrier permeability

Abstract

A QSAR model for predicting the blood brain barrier permeability (BBBP) in a large and heterogeneous variety of compounds (136 compounds) has been developed using approximate similarity (AS) matrices as predictors and PLS as multivariate regression technique. AS values fuse information of both the isomorphic similarity and nonisomorphic dissimilarity with the purpose of achieving an accurate predictive space. In addition to the fact of applying AS values to heterogeneous data sets, a new concept on graph isomorphism based on the extended maximum common subgraph (EMCS) is defined for the building of AS spaces considering the atoms and bonds, which are bridges between the isomorphic and nonisomorphic substructures. This new isomorphism detection has as objective to take into account the position and nature of the nucleus substituents, thus allowing the development of accurate models for large and diverse sets of compounds. After an outliers study, the training and test stages were made and the results obtained using several AS approaches were compared. Several validation processes were carried out by means of employing several test sets, and high predictive ability was obtained for all the cases (Q(2) = 0.81 and standard error in prediction, SEP = 0.29).

Computer modeling of blood brain barrier permeability of physiologically active compounds

“…Несмотря на обилие публикаций, до сегодняшнего дня не установлено четко, какова зависимость между величиной logBB и молекулярными дескрипторами. В подавляющем большинстве работ для построения QSAR моделей применяется линейная регрессия [90,91,93,110,117,118,122], а построенные модели основаны на использовании в качестве зависимых переменных различных физико-химических дескрипторов. Однако есть свидетельства и нелинейного характера такой зависимости [95,125,126].…”

Structural‐Similarity‐Based Approaches for the Development of Clustering and QSPR/QSAR Models in Chemical Databases

“…Recently, investigations have oriented their studies to combine different similarity approaches in order to obtain finer similarity values with an acceptable computational cost [36][37][38][39][40].…”

“…Approximate similarity (AS) combines graph-based and descriptor-based representations of the molecules in order to obtain a more accurate similarity measure [17,36,37,41]. The AS concept is based in considering the structural similarity as an incomplete measure of the resemblance between two molecules, so similarity only has into account common aspects of the molecular structures and the remaining structural fragments are not considered.…”

“…The AS approach has into account these nonisomorphic fragments (NIFs) [36,37] in order to build a new similarity metric combining graph-based and descriptor-based similarity approaches with the aim of obtaining more realistic similarity values: …”

A novel approach to spatial data fusion based on Bayesian network

“…Therefore, these have made the multi-sensor information fusion technology develop quickly [1][2]. In recent years, information fusion technology has developed with egregious speed, which extensively applied in many domains, including multi-objects detecting [3][4][5], identification [6][7][8], tracking [9][10][11] and the evaluation of battlefield scrutiny, situation and threaten, et al Meanwhile, information fusion has been gradually applied in other domains, such as intelligent vehicle, medical image treatment and diagnoses, weather forecast, earth science, agriculture, economy and commerce. Nowadays, information fusion plays more and more important role for data processing and information utilization.…”