Active meta-learning for predicting and selecting perovskite crystallization experiments.

Autonomous experimentation systems use algorithms and data from prior experiments to select and perform new experiments in order to meet a specified objective. In most experimental chemistry situations, there is a limited set of prior historical data available, and acquiring new data may be expensive and time consuming, which places constraints on machine learning methods. Active learning methods prioritize new experiment selection by using machine learning model uncertainty and predicted outcomes. Meta-learning methods attempt to construct models that can learn quickly with a limited set of data for a new task. In this paper, we applied the model-agnostic meta-learning (MAML) model and the Probabilistic LATent model for Incorporating Priors and Uncertainty in few-Shot learning (PLATIPUS) approach, which extends MAML to active learning, to the problem of halide perovskite growth by inverse temperature crystallization. Using a dataset of 1870 reactions conducted using 19 different organoammonium lead iodide systems, we determined the optimal strategies for incorporating historical data into active and meta-learning models to predict reaction compositions that result in crystals. We then evaluated the best three algorithms (PLATIPUS and active-learning k-nearest neighbor and decision tree algorithms) with four new chemical systems in experimental laboratory tests. With a fixed budget of 20 experiments, PLATIPUS makes superior predictions of reaction outcomes compared to other active-learning algorithms and a random baseline.

An accelerated background subtraction algorithm for processing high-resolution MS data and its application to metabolite identification.

BACKGROUND: Metabolite identification without radiolabeled compound is often challenging because of interference of matrix-related components. RESULTS: A novel and an effective background subtraction algorithm (A-BgS) has been developed to process high-resolution mass spectral data that can selectively remove matrix-related components. The use of a graphics processing unit with a multicore central processing unit enhanced processing speed several 1000-fold compared with a single central processing unit. A-BgS algorithm effectively removes background peaks from the mass spectra of biological matrices as demonstrated by the identification of metabolites of delavirdine and metoclopramide. CONCLUSION: The A-BgS algorithm is fast, user friendly and provides reliable removal of matrix-related ions from biological samples, and thus can be very helpful in detection and identification of in vivo and in vitro metabolites.