Rapid planning and analysis of high-throughput experiment arrays for reaction discovery

Abstract

High-throughput experimentation (HTE) is an increasingly important tool in reaction discovery. While the hardware for running HTE in the chemical laboratory has evolved significantly in recent years, there remains a need for software solutions to navigate data-rich experiments. Here we have developed phactor™, a software that facilitates the performance and analysis of HTE in a chemical laboratory. phactor™ allows experimentalists to rapidly design arrays of chemical reactions or direct-to-biology experiments in 24, 96, 384, or 1,536 wellplates. Users can access online reagent data, such as a chemical inventory, to virtually populate wells with experiments and produce instructions to perform the reaction array manually, or with the assistance of a liquid handling robot. After completion of the reaction array, analytical results can be uploaded for facile evaluation, and to guide the next series of experiments. All chemical data, metadata, and results are stored in machine-readable formats that are readily translatable to various software. We also demonstrate the use of phactor™ in the discovery of several chemistries, including the identification of a low micromolar inhibitor of the SARS-CoV-2 main protease. Furthermore, phactor™ has been made available for free academic use in 24- and 96-well formats via an online interface.

Fig. 1. Overview of the phactor™ software.

a Anatomy of a reaction as encoded by phactor™. b High-level software workflow of phactor™. Reaction arrays are designed from chemical inventories and reaction informatics. Resultant data is stored in delimited text (CSV) or in a relational database (SQLite3). phactor™ can convert results to Open Reaction Database (ORD) and Chemical Description Language (XDL) and is readily compatible with optimization programs such as EDBO+ and LabMate.ML. c Workflow of phactor™. Once the reaction array is designed, phactor™ provides human-readable or machine instructions to execute the dosing manually or robotically (UPLC ultra-performance liquid chromatography). d phactor™ supports custom volumes allowing for reaction arrays to be performed at any scale. At a minimum, the hardware needed to execute a reaction array is an autopipette and an array reactor block. e phactor™ facilitates the design and execution of ultraHTE in 1536 wellplates.

Fig. 2. Reaction arrays executed with the phactor™ software.

The reaction array design and results are shown here as displayed on phactor™. Colour bars adjacent to compound numbers correspond to the colour bars in the reaction array design grid generated by phactor™. Product/internal standard ratios are calculated using the observed UV-derived peak area, while assay yields account for differences in product absorptivity by calibrating to authentic samples of products. a Preliminary esterification hit leading to publication. b Optimized oxidative indolization conditions towards the synthesis of umifenovir. c Allylation catalyst/ligand concentration ratio and base reaction array analysed by conversion and selectivity.

Fig. 3. phactor™ has been utilized in a variety of synthetic campaigns.

a–i Chemistries discovered via reaction arrays designed with phactor™. All input and output files used to produce reaction arrays (a–f) are provided via an online repository in addition to all compiled HTE results in a machine-readable format. Reaction schemes can be found in the “Selected screening examples” section of the Supplementary Information.

Fig. 4. phactor™ facilitates ultraHTE direct-to-biology campaigns.

a Event workflow for performing ultraHTE using phactor™ and a Mosquito robot. b Design of 1280 well amide coupling plate. 80 amines were selected to react with carboxylic acid 26. Eight conditions were run in duplicate for each amine. c Results of the amide coupling are shown as a product/internal standard integration ratio from a 2-min LCMS injection of each well. The Mosquito robot is utilized to split the size 1536 plate into four sizes 384 plates for LCMS and bioassay analysis. d Percent inhibition of SARS-CoV-2 M^Pro when treated with a sample of the reaction mixture from the corresponding well. The 1280 plate is visually recreated. e IC₅₀ curves for three scaled-up compounds chosen from the reaction array. Compounds 27–29 display a range of assay and inhibitory responses.

Fig. 5. The six stages in the phactor™ workflow.

Each stage is progressed sequentially. With an input reagent CSV, reaction arrays can be designed in seconds. Once the experiment has been executed, a standardized output can be downloaded on the report stage.

Fig. 6. phactor™ enables rapid machine-learning analysis of multiple reaction arrays in tandem.

Standardized output files can be rapidly merged to create massive datasets. Shown is a tSNE (t-distributed stochastic neighbour embedding) of all products made in the decarboxylative–deaminative sp³–sp³ C–C coupling detailed in ref. , coloured by average product/internal standard.