Machine Learning of Interstellar Chemical Inventories

Abstract

The characterization of interstellar chemical inventories provides valuable insight into the chemical and physical processes in astrophysical sources. The discovery of new interstellar molecules becomes increasingly difficult as the number of viable species grows combinatorially, even when considering only the most thermodynamically stable. In this work, we present a novel approach for understanding and modeling interstellar chemical inventories by combining methodologies from cheminformatics and machine learning. Using multidimensional vector representations of molecules obtained through unsupervised machine learning, we show that identification of candidates for astrochemical study can be achieved through quantitative measures of chemical similarity in this vector space, highlighting molecules that are most similar to those already known in the interstellar medium. Furthermore, we show that simple, supervised learning regressors are capable of reproducing the abundances of entire chemical inventories, and predict the abundance of not yet seen molecules. As a proof-of-concept, we have developed and applied this discovery pipeline to the chemical inventory of a well-known dark molecular cloud, the Taurus Molecular Cloud 1 (TMC-1); one of the most chemically rich regions of space known to date. In this paper, we discuss the implications and new insights machine learning explorations of chemical space can provide in astrochemistry.

20 pages; 8 figures, 2 tables in the main text. 6 figures, 2 tables in the appendix. Accepted for publication in The Astrophysical Journal Letters. Molecule recommendations for TMC-1 can be found in the Zenodo repository: https://zenodo.org/record/5146276

References in corpus (16)

• Complex Chemistry in Star-Forming Regions: An Expanded Gas-Grain Warm-up Chemical Model
• Detection of Two Interstellar Polycyclic Aromatic Hydrocarbons via Spectral Matched Filtering
• The 2014 KIDA network for interstellar chemistry
• Discovery of the Pure Polycyclic Aromatic Hydrocarbon Indene ($c$-C${}_9$H${}_8$) with GOTHAM Observations of TMC-1
• Discovery of Interstellar Propylene (CH_2CHCH_3): Missing Links in Interstellar Gas-Phase Chemistry
• A New Reference Chemical Composition for TMC-1
• Early Science from GOTHAM: Project Overview, Methods, and the Detection of Interstellar Propargyl Cyanide (HCCCH${}_2$CN) in TMC-1
• Spatially resolved l-c3h+ emission in the horsehead photodissociation region: Further evidence for a top-down hydrocarbon chemistry
• Discovery of CH2CHCCH and detection of HCCN, HC4N, CH3CH2CN, and, tentatively, CH3CH2CCH in TMC-1
• Variable H${}^{13}$CO${}^{+}$ Emission in the IM Lup Disk: X-ray Driven Time-Dependent Chemistry?
• What Produces Dust Polarization in the HH 212 Protostellar Disk at 878 μm: Dust Self-Scattering or Dichroic Extinction?
• Detection of Interstellar HC${}_4$NC and an Investigation of Isocyanopolyyne Chemistry under TMC-1 Conditions
• Detection of Interstellar HC${}_5$O in TMC-1 with the Green Bank Telescope
• Deep K-band observations of TMC-1 with the Green Bank Telescope: Detection of HC7O, non-detection of HC11N, and a search for new organic molecules
• A 1.3 cm line survey toward IRC +10216
• Orion Source I's disk is salty

Cited by in corpus (5)

• A review of unsupervised learning in astronomy
• Astronomical Detection of the Interstellar Anion C10H- towards TMC-1 from the GOTHAM Large Program on the GBT
• Predicting binding energies of astrochemically relevant molecules via machine learning
• Implementation of Rare Isotopologues into Machine Learning of the Chemical Inventory of the Solar-Type Protostellar Source IRAS 16293-2422
• Machine learning-accelerated chemistry modeling of protoplanetary disks