A Deep Learning Approach to Antibiotic Discovery.

By Jonathan M. Stokes, Kevin Yang, Kyle Swanson, Wengong Jin, Andrés Cubillos-Ruiz, Nina M. Donghia, C. MacNair, S. French, L. Carfrae, Zohar Bloom-Ackermann and 10 more. 2020

1,250 citations recorded by the source corpus.

(or ciproﬂoxacin) were diluted into fresh LB at the equivalent of 1/10,000 of a dense culture, and once again introduced to varying concentrations of halicin (or ciproﬂoxacin) at two-fold serial dilutions. This procedure was performed every 24 hr over the course of 30 days. For spontaneous suppressor generation, (cid:1) 10 9 CFU of E. coli BW25113 grown in LB media was spread onto LB agar in 10 cm Petri dishes, either without antibiotics or supplemented with ciproﬂoxacin (Millipore Sigma) or halicin (TCI Chemicals) at the indicated concentrations. Plates were subsequently incubated at 37 (cid:4) C for seven days, at which time colonies from each plate were re-streaked onto LB and LB supplemented with antibiotics at the same concentration on which the colonies were originally grown. These plates were grown at 37 (cid:4) C overnight to monitor re-growth. For strain engineering, E. coli BW25113 D nsfA ::kan D nfsB ::cat was derived from BW25113 D nsfA ::kan via introduction of a cat gene to disrupt the nfsB ORF using the Lambda Red method (Datsenko and Wanner, 2000). Brieﬂy, 2 mL 2x YT media with BW25113 D nsfA

Subfields: Computational Theory and Mathematics, Molecular Biology, Pharmacology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Molecular Biology

In Pharmacology

Citations in the corpus

Relevance of Rotationally Equivariant Convolutions for Predicting Molecular Properties (2020, background): “The discovery of novel molecules has been accelerated by advances in computational quantum chemistry and machine learning assisted exploration of chemical space [1, 2, 3].”
Geometric deep learning on molecular representations (2021, background): “Since their introduction as a means to predict quantum chemical properties of small molecules at the density functional theory (DFT) level [5], GNNs have found many applications in quantum chemistry [68–72], drug discovery [37, 73, 74], CASP [75], and molecular property prediction [76, 77].”
GEOM: Energy-annotated molecular conformations for property prediction and molecular generation (2020, background): “aset contains 2,062 molecules that have been screened for growth inhibition of Pseudomonas aeruginosa (data accessed by request from [67]; 23 hits) and 1,580 molecules screened for E. Coli inhibition [1, 68] (57 hits). Secondary infections of COVID-19 patients can be caused by both of these pathogen”
Antibiotic resistance and persistence—Implications for human health and treatment perspectives (2020, neutral): “…antibacterial substances that are structurally different from conventional antibiotics and highlights the opportunities and impact machine learning can have on antibiotic discovery and prediction of properties of potential drugs while decreasing the cost of screening efforts (Stokes et al , 2020).”
Utilizing graph machine learning within drug discovery and development (2021, methodology): “With the design, synthesis and testing of de novo small molecules [48], the in vitro and in vivo testing of drug repurposing hypotheses [50], and target identification frameworks being conceptualized [47], we are potentially entering into a golden age of validation for GML within drug discovery and ”
Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development (2021, background): “Machine learning has already proved useful in the search of antibiotics (Stokes et al. 2020), polypharmacy (Zitnik, Agrawal & Leskovec 2018), drug repurposing for emerging diseases (Gysi et al. 2020), protein folding and design (Jumper et al. 2020, Gao et al. 2020), and biomolecular interactions…”
Deep learning in drug discovery: an integrative review and future challenges (2022, methodology): “The D-MPNN model was then employed by another research group to correctly forecast a kind of antibiotic named HALICIN, which demonstrated bactericide effects in models for mice (Stokes et al. 2020).”
Artificial Intelligence For Natural Product Drug Discovery (2023, methodology): “Other successful applications of bioactivity predictions have used representations such as 3D pharmacophore matching 57 of bio-activity signatures coupled to deep neural networks 63,64 .”