Citrate Controls Fe(II)-Catalyzed Transformation of Ferrihydrite by Complexation of the Labile Fe(III) Intermediate
Abstract not provided
Sheng, Anxu↗
SEARCH · Engineering Papers
Results for “affinity”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Transformation of Highly Stable Pt Single Sites on Defect Engineered Ceria into Robust Pt Clusters for Vehicle Emission Control
Engineering surface defects on metal oxide supports could help promote the dispersion of active sites and catalytic performance of supported catalysts. Herein, a strategy of ZrO 2 doping was proposed to create rich surface defects on CeO 2 (CZO) and, with these defects, to improve Pt dispersion and enhance its affinity as single sites to the CZO support (Pt/CZO). The strongly anchored Pt single sites on CZO support were initially not efficient for catalytic oxidation of CO/C 3 H 6 . However, after a simple activation by H 2 reduction, the catalytic oxidation performance over Pt/CZO catalyst was significantly boosted and better than Pt/CeO 2 . Pt/CZO catalyst also exhibited much higher thermal stability. The structural evolution of Pt active sites by H 2 treatment was systematically investigated on aged Pt/CZO and Pt/CeO 2 catalysts. With H 2 reduction, ionic Pt single sites were transformed into active Pt clusters. Much smaller Pt clusters were created on CZO (ca. 1.2 nm) than on CeO 2 (ca. 1.8 nm) due to stronger Pt-CeO 2 interaction on aged Pt/CZO. Consequently, more exposed active Pt sites were obtained on the smaller clusters surrounded by more oxygen defects and Ce 3+ species, which directly translated to the higher catalytic oxidation performance of activated Pt/CZO catalyst in vehicle emission control applications.
36 MATERIALS SCIENCE↗
Synergistic Enhancement of Lead and Selenate Uptake at the Barite (001)–Water Interface
The interactions of heavy metals with minerals influence the mobility and bioavailability of toxic elements in natural aqueous environments. The sorption of heavy metals on covalently bonded minerals is generally well described by surface complexation models (SCMs). However, understanding sorption on sparingly soluble minerals is challenging because of the dynamically evolving chemistry of sorbent surfaces. The interpretation can be even more complicated when multiple metal ions compete for sorption. In the present study, we observed synergistically enhanced uptake of lead and selenate on the barite (001) surface through two sorption mechanisms: lattice incorporation that dominates at lower coverages and two-dimensional monolayer growth that dominates at higher coverages. Further, we also observed a systematic increase in the sorption affinity with increasing co-sorbed ion coverages, different from the assumption of invariant binding constants for individual adsorption processes in classical SCMs. Computational simulations showed thermodynamically favorable co-incorporation of lead and selenate by simultaneously substituting for barium and sulfate in neighboring sites, resulting in the formation of molecular clusters that locally match the net dimension of the substrate lattice. These results emphasize the importance of ion–ion interactions at mineral–water interfaces that control the fate and transport of contaminants in the environment.
54 ENVIRONMENTAL SCIENCES↗
Complex Adsorption Behavior of Neodymium and Ytterbium on Structurally-Distinct Alumina Surfaces
New sources of rare earth elements (REEs) are needed to support a green energy transition. REEs adsorbed to aluminum-rich clays in weathering deposits represent important resources but the mechanisms responsible for their retention and ease of extraction are unresolved. Disordered coordination and co-occurrence of multiple species pose challenges to investigating REE adsorption processes via established spectroscopic methods. In this study, we applied element-specific surface crystallography methods to obtain a new perspective on the complexity of REE adsorption mechanisms and affinities. Alumina (001) and (012) crystal surfaces were utilized to evaluate surface-specific controls on Nd(III) and Yb(III) adsorption behavior. The REEs displayed similar total adsorption to alumina (001) as a mixture of inner- and outer-sphere complexes, but Nd displayed a greater proportion of inner-sphere binding. Adsorption of ordered inner- and outer-sphere REE species was substantially lower on alumina (012). These distinct behaviors reflect differences in the surface functional group charging and topography of the two surfaces. However, alumina (012) also hosted a substantial population of disordered adsorbed species, especially for Nd, potentially associated with Al vacancy surface defects. Here, the accumulation of light versus heavy REEs via adsorption in weathering deposits likely results from multiple, competing reactions affected by clay particle morphology. Leaching procedures for resource recovery should account for differential rates of desorption by coexisting inner- and outer-sphere REE surface complexes.
58 GEOSCIENCES↗
Multiple Sources of Riparian Wetland Suspended Solids during Episodic Rain Events: Influence on Uranium Transport
Suspended solids can be the primary vector for transporting contaminants in streams. The objective of this study was to determine whether changes in the properties of suspended solids during rain events impacted contaminant transport. Stream water was collected during five episodic events downstream from a U-contaminated wetland located in South Carolina, USA. The suspended particles were initially composed of Fe-flocs (particles formed in situ prior to the rain event) that had significantly greater Fe, Mn, organic-C, and U content than particles collected later during a sampling rain event. XANES and EXAFS revealed that U in the Fe-flocs was U(VI) and that it was not incorporated in a mineral structure but existed as inner- or outer-sphere adsorbed uranyl species associated with organic matter and Fe-oxides. The uranyl had an extraordinarily high affinity for the suspended solids, with solid to liquid U ratios of >72,000 (μg/kg)/(μg/L). After the initial flush of Fe-flocs, a greater fraction of the suspended solids had lower organic-C, Fe, Mn, and amorphous phases and were composed of more quartz, kaolinite, and gibbsite, resulting in lower U concentrations than those in the solids collected earlier in the rain event. This study highlights the importance of understanding suspended solids as transport vectors and their potential dynamic nature during rain events.
computer simulations↗
Impact of natural organic matter on plutonium vadose zone migration from an NH4Pu(V)O2CO3(s) source
Large amounts of plutonium produced at the Savannah River Site (SRS) in the latter half of the last century led to the release of plutonium to the subsurface. Low-level waste containing plutonium has also been disposed of in the SRS vadose zone. The mobility and risk posed by these plutonium sources is not well understood, in part, because of unknown interactions of the various mobile plutonium species with natural organic matter (NOM). This study investigated the influence of NOM on the behavior of plutonium(V) in the vadose zone through a combination of field lysimeter and laboratory studies. Well-defined solid sources of NH4PuVO2CO3(s) placed in 5-L lysimeters containing vadose zone soils NOM-amended or unamended were exposed to natural SRS meteorological conditions for three years. These lysimeters were removed from the field and analyzed using wet chemistry methods and X-ray absorption spectroscopy (XAS). Plutonium desorption from soils recovered from the lysimeters and laboratory batch sorption experiments using Pu-free soils both indicated that surface mediated reduction of Pu(V) to Pu(IV) was influencing Pu sorption behavior. For both lysimeters (amended and unamended with NOM), Pu migrated slowly through the lysimeter core, with > 95% of the Pu remaining within 2 cm of the source. However, without the NOM amendments, Pu was transported significantly farther than in the presence of NOM. Thus, the presence of NOM appeared to decrease the mobility. Working hypotheses to explain this behavior are that 1) formation of ternary surface-NOM-Pu complexes enhanced Pu sorption affinity causing less mobility and/or 2) formation of these ternary surface-NOM-Pu complexes prevented reoxidaiton of Pu(IV) to more mobile Pu(V) in the pore water.
plutonium, organic matter↗
Porous Organic Cages CC3 and CC2 as Adsorbents for the Separation of Carbon Dioxide from Nitrogen and Hydrogen
The selective capture of carbon dioxide over nitrogen and hydrogen is of great industrial interest in flue gas and hydrogen purification respectively. Microporous adsorbents are highly suitable materials to preferentially adsorb gases. Here, in particular, Porous organic cages (POCs) with tunable hierarchically ordered micropores, high surface area, and thermodynamic affinity for CO 2 make them appealing candidates for these applications. Herein, we demonstrate that two prototypical POCs denoted as CC3, and CC2 with limiting pore aperture of 3.6 Å can selectively separate CO 2 from N 2 , and H 2 . For CC3 adsorption selectivities as high as ~ 8 and ~20 for CO 2 /N 2 and CO 2 /H 2 respectively were observed. For CC2 adsorption selectivities as high as ~ 9 and ~35 for CO 2 /N 2 and CO 2 /H 2 respectively were observed. Interestingly, the adsorption selectivity of the studied gases correlated linearly with polarizability selectivity.
Carbon Dioxide↗
Performance Degradation of Amine-Infused Fiber Sorbents for Direct Air Capture: Mechanisms and Solutions
Sorbent stability poses significant impacts on longterm performance of direct air capture (DAC) of CO 2 and levelized cost of capture (LCOC). We report the DAC performance degradation of amine-infused fiber sorbents based on poly(ethylenimine) (PEI), mesoporous SiO 2 , and cellulose acetate (CA) over CO 2 cyclic sorption cycles in a nonoxidative environment. Infrared and nuclear magnetic resonance spectra indicate that the aminolysis reactions between CA ester moieties and PEI amine sites lead to the formation of acetamides and hence lower CO 2 affinities of the sorbents. This stability issue can be remedied by hydrolysis treatment of the CA fiber sorbents before PEI impregnation or replacing CA with poly(ether sulfone). This study underscores the importance of selecting proper support or additive materials of DAC contactors that are compatible with active species of CO 2 capture.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
Stability of Metal–Organic Framework-Supported Amines under Exposure to Ozone Generated from Air
Amine compounds supported on porous materials such as metal–organic frameworks (MOFs) have shown promising performance for direct air capture (DAC) due to their enhanced affinity for CO 2 . Although features such as adsorption capacity and selectivity are paramount in these composites, their long-term stability has a major impact on the operating cost of DAC systems. In this work, changes in carbon capture performance, crystallinity, porosity and chemical environment of the constituting atoms of MOF-amine composites are explored after exposure to ozone and NOx impurities generated from corona discharge applied to air. From the obtained results, the stabilities of Mg 2 (dobpdc) (dobpdc 4– = 4,4′-dioxidobiphenyl-3,3′-dicarboxylate) grafted with ethylenediamine (en), N-methylethylenediamine (men), and N,N-dimethylethylenediamine (dmen), as well as MIL-101(Cr) MOF impregnated with polyethylenimine (PEI), are compared. A negative effect in the overall CO 2 adsorption capacity is observed for all MOF composites after exposure, as well as a decrease in the adsorption step pressure of CO 2 for Mg 2 (dobpdc) amine-grafted composites, as shown via dynamic gravimetric adsorption experiments. Spectroscopic analyses indicate that oxidation of amine groups through the formation of nitro functional groups occurs as well as a decrease in the electron-donation interaction between the supported amines and the metal nodes of the MOFs.
adsorption↗
An Aluminum-Based Metal–Organic Cage for Cesium Capture
Metal–organic cages are a class of supramolecular structures that often require the careful selection of organic linkers and metal nodes. Of this class, few examples of metal–organic cages exist where the nodes are composed of main group metals. Herein, we have prepared an aluminum-based metal–organic cage, H 8 [Al 8 (pdc) 8 (OAc) 8 O 4 ] (Al-pdc-AA), using inexpensive and commercially available materials. The cage formation was achieved via solvothermal self-assembly of solvated aluminum and pyridine-dicarboxylic linkers in the presence of a capping agent, acetic acid. The obtained supramolecular structure was characterized by single-crystal X-ray diffraction (SCXRD), thermogravimetric analysis, and NMR spectroscopy. Based on crystal structure and computational analyses, the cage has a 3.7 Å diameter electron-rich cavity suitable for the binding of cations such as cesium (ionic radius of 1.69 Å). Here, the host–guest interactions were probed with 1 H and 133 Cs NMR spectroscopy in DMSO, where at low concentrations, Cs + binds to Al-pdc-AA in a 1:1 ratio. The binding site was identified from the crystal structure of CsH 7 [Al 8 (pdc) 8 (OAc) 8 O 4 ] (Cs + Al-pdc-AA), and a binding affinity of ~10 6 –10 7 M –1 was determined from NMR titration experiments. The Al-pdc-AA showed improved selectivity for cesium binding over alkali metal cations (Cs + > Rb + > K + >> Na + ~ Li + ). Collectively, the study reports a novel aluminum cage that can serve as a promising host for efficient and selective cesium removal.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
Chelating Rare-Earth Metals (Ln 3+ ) and 225 Ac 3+ with the Dual-Size-Selective Macrocyclic Ligand Py 2 -Macrodipa
Radioisotopes of metallic elements, or radiometals, are widely employed in both therapeutic and diagnostic nuclear medicine. For this application, chelators that efficiently bind the radiometal of interest and form a stable metal–ligand complex with it are required. Toward the development of new chelators for nuclear medicine, we recently reported a novel class of 18-membered macrocyclic chelators that is characterized by their ability to form stable complexes with both large and small rare-earth metals (Ln 3+ ), a property referred to as dual size selectivity. A specific chelator in this class called py-macrodipa, which contains one pyridyl group within its macrocyclic core, was established as a promising candidate for 135 La 3+ , 213 Bi 3+ , and 44 Sc 3+ chelation. Building upon this prior work, here we report the synthesis and characterization of a new chelator called py 2 -macrodipa with two pyridyl units fused into the macrocyclic backbone. Its coordination chemistry with the Ln 3+ series was investigated by NMR spectroscopy, X-ray crystallography, density functional theory (DFT) calculations, analytical titrations, and transchelation assays. These studies reveal that py 2 -macrodipa retains the expected dual size selectivity and possesses an enhanced thermodynamic affinity for all Ln 3+ compared to py-macrodipa. By contrast, the kinetic stability of Ln 3+ complexes with py 2 -macrodipa is only improved for the light, large Ln 3+ ions. Based upon these observations, we further assessed the suitability of py 2 -macrodipa for use with 225 Ac 3+ , a large radiometal with valuable properties for targeted α therapy. Radiolabeling and stability studies revealed py 2 -macrodipa to efficiently incorporate 225 Ac 3+ and to form a complex that is inert in human serum over 3 weeks. Although py 2 -macrodipa does not surpass the state-of-the-art chelator macropa for 225 Ac 3+ chelation, it does provide another effective 225 Ac 3+ chelator. Furthermore, these studies shed light on the fundamental coordination chemistry of the Ln 3+ series and may inspire future chelator design efforts.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
ATCUN-like Copper Site in βB2-Crystallin Plays a Protective Role in Cataract-Associated Aggregation
Cataract is the leading cause of blindness worldwide, and it is caused by crystallin damage and aggregation. Senile cataractous lenses have relatively high levels of metals, while some metal ions can directly induce the aggregation of human γ-crystallins. Here, for this work, we evaluated the impact of divalent metal ions in the aggregation of human βB2-crystallin, one of the most abundant crystallins in the lens. Turbidity assays showed that Pb 2+ , Hg 2+ , Cu 2+ , and Zn 2+ ions induce the aggregation of βB2-crystallin. Metal-induced aggregation is partially reverted by a chelating agent, indicating the formation of metal-bridged species. Our study focused on the mechanism of copper-induced aggregation of βB2-crystallin, finding that it involves metal-bridging, disulfide-bridging, and loss of protein stability. Circular dichroism and electron paramagnetic resonance (EPR) revealed the presence of at least three Cu 2+ binding sites in βB2-crystallin, one of them with spectroscopic features typical for Cu 2+ bound to an amino-terminal copper and nickel (ATCUN) binding motif, which is found in Cu transport proteins. The ATCUN-like Cu binding site is located at the unstructured N-terminus of βB2-crystallin, and it could be modeled by a peptide with the first six residues in the protein sequence (NH 2 -ASDHQF-). Isothermal titration calorimetry indicates a nanomolar Cu 2+ binding affinity for the ATCUN-like site. An N-truncated form of βB2-crystallin is more susceptible to Cu-induced aggregation and is less thermally stable, indicating a protective role for the ATCUN-like site. EPR and X-ray absorption spectroscopy studies reveal the presence of a copper redox active site in βB2-crystallin that is associated with metal-induced aggregation and formation of disulfide-bridged oligomers. Our study demonstrates metal-induced aggregation of βB2-crystallin and the presence of putative copper binding sites in the protein. Whether the copper-transport ATCUN-like site in βB2-crystallin plays a functional/protective role or constitutes a vestige from its evolution as a lens structural protein remains to be elucidated.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
Influence of Aqueous Phase Acidity on Ln(III) Coordination by N , N , N ', N '-Tetraoctyldiglycolamide
Here, this study highlights the importance of combining distribution ratio measurements with multiple spectroscopic techniques to provide a more comprehensive understanding of organic phase Ln coordination chemistry. Solvent extraction investigations with N,N,N',N'-tetraoctyldiglycolamide (TODGA) in n-heptane reveal the sensitivity of Ln complexation to the HNO 3 concentration. Distribution ratio measurements in tandem with UV–Vis demonstrated that increasing the concentration of HNO3 above 0.5 M with a constant NO 3 – of 1 M increases the number of coordinating TODGA molecules, from a 1:2 to a 1:3 Ln:TODGA complex. At each concentration of HNO 3 considered herein (from 0.01 to 1 M), Eu lifetime analysis demonstrated no evidence of H 2 O coordination. Results from Fourier transform infrared investigations suggest the presence of inner-sphere NO 3 – under low concentrations of HNO 3 when the 1:2 Ln:TODGA complex is present. Increasing the HNO 3 concentration above 0.5 M increases the propensity for outer-sphere interactions by removing the coordinated NO 3 – and saturating the Ln coordination sphere with three TODGA molecules, resulting in the well-established cationic, trischelate homoleptic [Ln(TODGA) 3 ] 3+ complex. This work demonstrates the importance in considering the NO 3 – source for solvent extraction systems. In particular, for systems with an affinity for outer-sphere interactions with molar concentrations of HNO 3 , changing the NO 3 – source can change the inner-sphere coordination of the Ln complex, which, in turn, affects the separation efficacy.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
Selective Lanthanide Sensing with Gold Nanoparticles and Hydroxypyridinone Chelators
The octadentate hydroxypyridinone chelator 3,4,3-LI(1,2-HOPO) is a promising therapeutic agent because of its high affinity for f-block elements and noncytotoxicity at medical dosages. The interaction between 3,4,3-LI(1,2-HOPO) and other biomedically relevant metals such as gold, however, has not been explored. Gold nanoparticles functionalized with chelators have demonstrated great potential in theranostics, yet thus far, no protocol that combines 3,4,3-LI(1,2-HOPO) and colloidal gold has been developed. Here, we characterize the solution thermodynamic properties of the complexes formed between 3,4,3-LI(1,2-HOPO) and Au 3+ ions and demonstrate how under specific pH conditions the chelator promotes the growth of gold nanoparticles, acting as both reducing and stabilizing agent. 3,4,3-LI(1,2-HOPO) ligands on the nanoparticle surface remain active and selective toward f-block elements, as evidenced by gold nanoparticle selective aggregation. Finally, a new colorimetric assay capable of reaching the detection levels necessary for the quantification of lanthanides in waste from industrial processes is developed based on the inhibition of particle growth by lanthanides.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
Ensemble Simulations and Experimental Free Energy Distributions: Evaluation and Characterization of Isoxazole Amides as SMYD3 Inhibitors
Optimization of binding affinities for ligands to their target protein is a primary objective in rational drug discovery. Herein, we report on a collaborative study that evaluates various compounds designed to bind to the SET and MYND domain-containing protein 3 (SMYD3). SMYD3 is a histone methyltransferase and plays an important role in transcriptional regulation in cell proliferation, cell cycle, and human carcinogenesis. Experimental measurements using the scintillation proximity assay show that the distributions of binding free energies from a large number of independent measurements exhibit non-normal properties. We use ESMACS (enhanced sampling of molecular dynamics with approximation of continuum solvent) and TIES (thermodynamic integration with enhanced sampling) protocols to predict the binding free energies and to provide a detailed chemical insight into the nature of ligand-protein binding. Our results show that the 1-trajectory ESMACS protocol works well for the set of ligands studied here. Although one unexplained outlier exists, we obtain excellent statistical ranking across the set of compounds from the ESMACS protocol and good agreement between calculations and experiments for the relative binding free energies from the TIES protocol. ESMACS and TIES are again found to be powerful protocols for the accurate comparison of the binding free energies.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
Identification of Small-Molecule Inhibitors of Fibroblast Growth Factor 23 Signaling via In Silico Hot Spot Prediction and Molecular Docking to α-Klotho
Fibroblast growth factor 23 (FGF23) is a therapeutic target for treating hereditary and acquired hypophosphatemic disorders, such as X-linked hypophosphatemic (XLH) rickets and tumor-induced osteomalacia (TIO), respectively. FGF23-induced hypophosphatemia is mediated by signaling through a ternary complex formed by FGF23, the FGF receptor (FGFR), and α-Klotho. Currently, disorders of excess FGF23 are treated with an FGF23-blocking antibody, burosumab. Small-molecule drugs that disrupt protein/protein interactions necessary for the ternary complex formation offer an alternative to disrupting FGF23 signaling. In this study, the FGF23:α-Klotho interface was targeted to identify small-molecule protein/protein interaction inhibitors since it was computationally predicted to have a large fraction of hot spots and two druggable residues on α-Klotho. We further identified Tyr433 on the KL1 domain of α-Klotho as a promising hot spot and α-Klotho as an appropriate drug-binding target at this interface. Subsequently, we performed in silico docking of ~5.5 million compounds from the ZINC database to the interface region of α-Klotho from the ternary crystal structure. Following docking, 24 and 20 compounds were in the final list based on the lowest binding free energies to α-Klotho and the largest number of contacts with Tyr433, respectively. Herein, five compounds were assessed experimentally by their FGF23-mediated extracellular signal-regulated kinase (ERK) activities in vitro, and two of these reduced activities significantly. Both these compounds were predicted to have favorable binding affinities to α-Klotho but not have a large number of contacts with the hot spot Tyr433. ZINC12409120 was found experimentally to disrupt FGF23:α-Klotho interaction to reduce FGF23-mediated ERK activities by 70% and have a half maximal inhibitory concentration (IC50) of 5.0 ± 0.23 μM. Molecular dynamics (MD) simulations of the ZINC12409120:α-Klotho complex starting from in silico docking poses reveal that the ligand exhibits contacts with residues on the KL1 domain, the KL1–KL2 linker, and the KL2 domain of α-Klotho simultaneously, thereby possibly disrupting the regular function of α-Klotho and impeding FGF23:α-Klotho interaction. ZINC12409120 is a candidate for lead optimization.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
AI-Accelerated Design of Targeted Covalent Inhibitors for SARS-CoV-2
Direct-acting antivirals for the treatment of the COVID-19 pandemic caused by the SARS-CoV-2 virus are needed to complement vaccination efforts. Given the ongoing emergence of new variants, automated experimentation, and active learning based fast workflows for antiviral lead discovery remain critical to our ability to address the pandemic’s evolution in a timely manner. While several such pipelines have been introduced to discover candidates with noncovalent interactions with the main protease (M pro ), here we developed a closed-loop artificial intelligence pipeline to design electrophilic warhead-based covalent candidates. Here, this work introduces a deep learning-assisted automated computational workflow to introduce linkers and an electrophilic “warhead” to design covalent candidates and incorporates cutting-edge experimental techniques for validation. Using this process, promising candidates in the library were screened, and several potential hits were identified and tested experimentally using native mass spectrometry and fluorescence resonance energy transfer (FRET)-based screening assays. We identified four chloroacetamide-based covalent inhibitors of M pro with micromolar affinities (K I of 5.27 μM) using our pipeline. Experimentally resolved binding modes for each compound were determined using room-temperature X-ray crystallography, which is consistent with the predicted poses. The induced conformational changes based on molecular dynamics simulations further suggest that the dynamics may be an important factor to further improve selectivity, thereby effectively lowering KI and reducing toxicity. These results demonstrate the utility of our modular and data-driven approach for potent and selective covalent inhibitor discovery and provide a platform to apply it to other emerging targets.
60 APPLIED LIFE SCIENCES↗
Ligand-Based Compound Activity Prediction via Few-Shot Learning
Predicting the activities of new compounds against biophysical or phenotypic assays based on the known activities of one or a few existing compounds is a common goal in early stage drug discovery. This problem can be cast as a “few-shot learning” challenge, and prior studies have developed few-shot learning methods to classify compounds as active versus inactive. However, the ability to go beyond classification and rank compounds by expected affinity is more valuable. We describe Few-Shot Compound Activity Prediction (FS-CAP), a novel neural architecture trained on a large bioactivity data set to predict compound activities against an assay outside the training set, based on only the activities of a few known compounds against the same assay. Our model aggregates encodings generated from the known compounds and their activities to capture assay information and uses a separate encoder for the new compound whose activity is to be predicted. The new method provides encouraging results relative to traditional chemical-similarity-based techniques as well as other state-of-the-art few-shot learning methods in tests on a variety of ligand-based drug discovery settings and data sets.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗