${k}$ nearest neighbor (kNN) method is a popular classification method in data mining and statistics because of its simple implementation and significant classification performance. However, it is impractical for traditional kNN methods to assign a fixed ${k}$ value (even though set by experts) to all test samples. Previous solutions assign different $k$ values to different test samples by the cross validation method but are usually time-consuming. This paper proposes a kTree method to learn different optimal $k$ values for different test/new samples, by involving a training stage in the kNN classification. Specifically, in the training stage, kTree method first learns optimal $k$ values for all training samples by a new sparse reconstruction model, and then constructs a decision tree (namely, kTree) using training samples and the learned optimal $k$ values. In the test stage, the kTree fast outputs the optimal $k$ value for each test sample, and then, the kNN classification can be conducted using the learned optimal $k$ value and all training samples. As a result, the proposed kTree method has a similar running cost but higher classification accuracy, compared with traditional kNN methods, which assign a fixed ${k}$ value to all test samples. Moreover, the proposed kTree method needs less running cost but achieves similar classification accuracy, compared with the newly kNN methods, which assign different ${k}$ values to different test samples. This paper further proposes an improvement version of kTree method (namely, k*Tree method) to speed its test stage by extra storing the information of the training samples in the leaf nodes of kTree, such as the training samples located in the leaf nodes, their kNNs, and the nearest neighbor of these kNNs. We call the resulting decision tree as k*Tree, which enables to conduct kNN classification using a subset of the training samples in the leaf nodes rather than all training samples used in the newly kNN methods. This actually reduces running cost of test stage. Finally, the experimental results on 20 real data sets showed that our proposed methods (i.e., kTree and k*Tree) are much more efficient than the compared methods in terms of classification tasks.

Efficient kNN Classification With Different Numbers of Nearest Neighbors

Machine learning in chemoinformatics and drug discovery.

The n-octanol/water partition coefficient (log Po/w) is a key physicochemical parameter for drug discovery, design, and development. Here, we present a physics-based approach that shows a strong linear correlation between the computed solvation free energy in implicit solvents and the experimental log Po/w on a cleansed data set of more than 17,500 molecules. After internal validation by five-fold cross-validation and data randomization, the predictive power of the most interesting multiple linear model, based on two GB/SA parameters solely, was tested on two different external sets of molecules. On the Martel druglike test set, the predictive power of the best model (N = 706, r = 0.64, MAE = 1.18, and RMSE = 1.40) is similar to six well-established empirical methods. On the 17-drug test set, our model outperformed all compared empirical methodologies (N = 17, r = 0.94, MAE = 0.38, and RMSE = 0.52). The physical basis of our original GB/SA approach together with its predictive capacity, computational effi...

iLOGP: A Simple, Robust, and Efficient Description of n-Octanol/Water Partition Coefficient for Drug Design Using the GB/SA Approach.

The annotation of small molecules remains a major challenge in untargeted mass spectrometry-based metabolomics. We here critically discuss structured elucidation approaches and software that are designed to help during the annotation of unknown compounds. Only by elucidating unknown metabolites first is it possible to biologically interpret complex systems, to map compounds to pathways and to create reliable predictive metabolic models for translational and clinical research. These strategies include the construction and quality of tandem mass spectral databases such as the coalition of MassBank repositories and investigations of MS/MS matching confidence. We present in silico fragmentation tools such as MS-FINDER, CFM-ID, MetFrag, ChemDistiller and CSI:FingerID that can annotate compounds from existing structure databases and that have been used in the CASMI (critical assessment of small molecule identification) contests. Furthermore, the use of retention time models from liquid chromatography and the utility of collision cross-section modelling from ion mobility experiments are covered. Workflows and published examples of successfully annotated unknown compounds are included.

/pdf/software-tools-and-approaches-for-compound-identification-of-3zka7g94zo.pdf

Software Tools and Approaches for Compound Identification of LC-MS/MS Data in Metabolomics.

The K Nearest Neighbor (kNN) method has widely been used in the applications of data mining and machine learning due to its simple implementation and distinguished performance. However, setting all test data with the same k value in the previous kNN methods has been proven to make these methods impractical in real applications. This article proposes to learn a correlation matrix to reconstruct test data points by training data to assign different k values to different test data points, referred to as the Correlation Matrix kNN (CM-kNN for short) classification. Specifically, the least-squares loss function is employed to minimize the reconstruction error to reconstruct each test data point by all training data points. Then, a graph Laplacian regularizer is advocated to preserve the local structure of the data in the reconstruction process. Moreover, an e1-norm regularizer and an e2, 1-norm regularizer are applied to learn different k values for different test data and to result in low sparsity to remove the redundant/noisy feature from the reconstruction process, respectively. Besides for classification tasks, the kNN methods (including our proposed CM-kNN method) are further utilized to regression and missing data imputation. We conducted sets of experiments for illustrating the efficiency, and experimental results showed that the proposed method was more accurate and efficient than existing kNN methods in data-mining applications, such as classification, regression, and missing data imputation.

https://dl.acm.org/doi/pdf/10.1145/2990508

Learning k for kNN Classification

One of the OECD principles for model validation requires defining the Applicability Domain (AD) for the QSAR models. This is important since the reliable predictions are generally limited to query chemicals structurally similar to the training compounds used to build the model. Therefore, characterization of interpolation space is significant in defining the AD and in this study some existing descriptor-based approaches performing this task are discussed and compared by implementing them on existing validated datasets from the literature. Algorithms adopted by different approaches allow defining the interpolation space in several ways, while defined thresholds contribute significantly to the extrapolations. For each dataset and approach implemented for this study, the comparison analysis was carried out by considering the model statistics and relative position of test set with respect to the training space.

/pdf/comparison-of-different-approaches-to-define-the-1csx4x046s.pdf

Comparison of different approaches to define the applicability domain of QSAR models.

With the growing popularity of using QSAR predictions towards regulatory purposes, such predictive models are now required to be strictly validated, an essential feature of which is to have the model’s Applicability Domain (AD) defined clearly. Although in recent years several different approaches have been proposed to address this goal, no optimal approach to define the model’s AD has yet been recognized. This study proposes a novel descriptor-based AD method which accounts for the data distribution and exploits k-Nearest Neighbours (kNN) principle to derive a heuristic decision rule. The proposed method is a three-stage procedure to address several key aspects relevant in judging the reliability of QSAR predictions. Inspired from the adaptive kernel method for probability density function estimation, the first stage of the approach defines a pattern of thresholds corresponding to the various training samples and these thresholds are later used to derive the decision rule. Criterion deciding if a given test sample will be retained within the AD is defined in the second stage of the approach. Finally, the last stage tries reflecting upon the reliability in derived results taking model statistics and prediction error into account. The proposed approach addressed a novel strategy that integrated the kNN principle to define the AD of QSAR models. Relevant features that characterize the proposed AD approach include: a) adaptability to local density of samples, useful when the underlying multivariate distribution is asymmetric, with wide regions of low data density; b) unlike several kernel density estimators (KDE), effectiveness also in high-dimensional spaces; c) low sensitivity to the smoothing parameter k; and d) versatility to implement various distances measures. The results derived on a case study provided a clear understanding of how the approach works and defines the model’s AD for reliable predictions.

/pdf/defining-a-novel-k-nearest-neighbours-approach-to-assess-the-3fewgqe42p.pdf

Defining a novel k-nearest neighbours approach to assess the applicability domain of a QSAR model for reliable predictions

Locally centred Mahalanobis distance: a new distance measure with salient features towards outlier detection.

Several classical and two recently proposed Applicability Domain (AD) approaches were implemented on a set of three classification models retrieved from a published study to assess the ready biodegradability of chemicals. Each model was associated with an optimal AD approach based on its ability to a) retain maximum test molecules within the model's AD, b) be appropriate for the strategy used towards model development and c) show reasonably converging results with those derived with other AD approaches used. A decision criterion was also set to evaluate the AD of two consensus models that were developed in the original study. An overview of test molecules excluded from the AD of all the five biodegradability models was provided including an attempt to identify the major structural features and molecular descriptors possibly relevant in deciding upon their ready biodegradability. Apart from the test set, an overview of the results derived on the external validation set molecules was provided.

F Sahigara

Papers

Comparison of different approaches to define the applicability domain of QSAR models.

Defining a novel k-nearest neighbours approach to assess the applicability domain of a QSAR model for reliable predictions

Locally centred Mahalanobis distance: a new distance measure with salient features towards outlier detection.

Assessing the validity of QSARs for ready biodegradability of chemicals: an applicability domain perspective.

Comparison of approaches to define Applicability Domain for the application of QSAR models