Search for: All records

Most molecular diagram parsers recover chemical structure from raster images (e.g., PNGs). However, many PDFs include commands giving explicit locations and shapes for characters, lines, and polygons. We present a new parser that uses these born-digital PDF primitives as input. The parsing model is fast and accurate, and does not require GPUs, Optical Character Recognition (OCR), or vectorization. We use the parser to annotate raster images and then train a new multi-task neural network for recognizing molecules in raster images.We evaluate our parsers using SMILES and standard benchmarks, along with a novel evaluation protocol comparing molecular graphs directly that supports automatic error compilation and reveals errors missed by SMILES-based evaluation. On the synthetic USPTObenchmark, our born-digital parser obtains a recognition rate of 98.4% (1% higher than previous models) and our relatively simple neural parser for raster images obtains a rate of 85% using less training data than existing neural approaches (thousands vs. millions of molecules). 

The management and analysis of large in silico molecular libraries is pivotal in many areas of modern chemistry. The adoption and success of data-oriented approaches to chemical research is dependent on the ease of handling large collections of in silico molecular structures in a programmatic way. Herein, we introduce the MOLecular LIibrary toolkit, “molli”, which is a Python 3 chemoinformatics module that provides a streamlined interface for manipulating large in silico libraries. Three-dimensional, combinatorial molecule libraries can be expanded directly from two-dimensional chemical structure fragments stored in CDXML files with high stereochemical fidelity. Geometry optimization, property calculation, and conformer generation are executed by interfacing with widely used computational chemistry programs such as OpenBabel, RDKit, ORCA, and xTB/CREST. Conformer-dependent grid-based feature calculators provide numerical representation suitable for diversity analysis, and interface to robust three-dimensional visualization tools provide comprehensive images to enhance human understanding of libraries with thousands of members. The package includes command-line interface in addition to Python classes to streamline frequently used workflows. This work describes the development and implementation of molli 1.0 and highlights the available functionality. Parallel performance is benchmarked on various hardware platforms and common workflows are demonstrated for different tasks ranging from optimized grid-based descriptor calculation on catalyst libraries to NMR prediction workflow from CDXML files.