Fr\'echet ChemblNet Distance: A metric for generative models for molecules
نویسندگان
چکیده
The new wave of successful generative models in machine learning has increased the interest in deep learning driven de novo drug design. However, assessing the performance of such generative models is notoriously difficult. Metrics that are typically used to assess the performance of such generative models are the percentage of chemically valid molecules or the similarity to real molecules in terms of particular descriptors, such as the partition coefficient (logP) or druglikeness. However, method comparison is difficult because of the inconsistent use of evaluation metrics, the necessity for multiple metrics, and the fact that some of these measures can easily be tricked by simple rule-based systems. We propose a novel distance measure between two sets of molecules, called Fréchet ChemblNet distance (FCD), that can be used as an evaluation metric for generative models. The FCD is similar to a recently established performance metric for comparing image generation methods, the Fréchet Inception Distance (FID). Whereas the FID uses one of the hidden layers of InceptionNet, the FCD utilizes the penultimate layer of a deep neural network called “ChemblNet”, which was trained to predict drug activities. Thus, the FCD metric takes into account chemically and biologically relevant information about molecules, and also measures the diversity of the set via the distribution of generated molecules. The FCD’s advantage over previous metrics is that it can detect if generated molecules are a) diverse and have similar b) chemical and c) biological properties as real molecules. We further provide an easy-to-use implementation that only requires the SMILES representation of the generated molecules as input to calculate the FCD. Implementations are available at: github.com/bioinf-jku/FCD.
منابع مشابه
Metric Learning-based Generative Adversarial Network
Generative Adversarial Networks (GANs), as a framework for estimating generative models via an adversarial process, have attracted huge attention and have proven to be powerful in a variety of tasks. However, training GANs is well known for being delicate and unstable, partially caused by its sigmoid cross entropy loss function for the discriminator. To overcome such a problem, many researchers...
متن کاملA CHARACTERIZATION FOR METRIC TWO-DIMENSIONAL GRAPHS AND THEIR ENUMERATION
The textit{metric dimension} of a connected graph $G$ is the minimum number of vertices in a subset $B$ of $G$ such that all other vertices are uniquely determined by their distances to the vertices in $B$. In this case, $B$ is called a textit{metric basis} for $G$. The textit{basic distance} of a metric two dimensional graph $G$ is the distance between the elements of $B$. Givi...
متن کاملThe Existence Theorem for Contractive Mappings on $wt$-distance in $b$-metric Spaces Endowed with a Graph and its Application
In this paper, we study the existence and uniqueness of fixed points for mappings with respect to a $wt$-distance in $b$-metric spaces endowed with a graph. Our results are significant, since we replace the condition of continuity of mapping with the condition of orbitally $G$-continuity of mapping and we consider $b$-metric spaces with graph instead of $b$-metric spaces, under which can be gen...
متن کاملThe Banach Type Contraction for Mappings on Algebraic Cone Metric Spaces Associated with An Algebraic Distance and Endowed with a Graph
In this work, we define the notion of an algebraic distance in algebraic cone metric spaces defined by Niknam et al. [A. Niknam, S. Shamsi Gamchi and M. Janfada, Some results on TVS-cone normed spaces and algebraic cone metric spaces, Iranian J. Math. Sci. Infor. 9 (1) (2014), 71--80] and introduce some its elementary properties. Then we prove the existence and uniqueness of fixed point for a B...
متن کاملCommon fixed points for a pair of mappings in $b$-Metric spaces via digraphs and altering distance functions
In this paper, we discuss the existence and uniqueness of points of coincidence and common fixed points for a pair of self-mappings satisfying some generalized contractive type conditions in $b$-metric spaces endowed with graphs and altering distance functions. Finally, some examples are provided to justify the validity of our results.
متن کامل