Unknown Transformation Research Articles

Hydroxyl radical oxidation of chemical contaminants on indoor surfaces and dust

Humans are widely exposed to semivolatile organic contaminants in indoor environments. Many contaminants have long lifetimes following partitioning to the large surface reservoirs present indoors, which leads to long exposure times to gas-phase oxidants and multiphase chemistry. Studies have shown selective multiphase oxidation of organics on indoor surfaces, but the presence of hydroxyl radicals with nonselective reactivity and evidence of multiphase OH radical reactivity toward common indoor contaminants indicates that there may be additional unknown transformation chemistry indoors. We screened genuine indoor samples for 60 OH radical oxidation products of the common plasticizer and endocrine-disrupting contaminant bis(2-ethylhexyl) phthalate (DEHP) identified in laboratory experiments using nontargeted high-resolution mass spectrometry. At least 30 and 10 of these products are observed in indoor dust and DEHP films exposed to ambient indoor conditions, respectively, indicating that multiphase OH reactions occur indoors. Using the PROTEX model and a multimedia indoor chemical fate model, we demonstrate that these products have long indoor lifetimes and cause a higher potential for human exposure than DEHP. Some of these products are more active endocrine disruptors than DEHP itself, but most have unknown toxicities. Coexposure to all oxidation products will likely have an additive effect, leading to higher human health risks from indoor organic contaminants than previously thought. Due to the nonselective reactivity of OH radicals, it is likely that most indoor contaminants follow similar chemistry, and further study is needed to understand the prevalence and human health implications of such multiphase chemistry.

Modelling multivariate spatio-temporal data with identifiable variational autoencoders

Modelling multivariate spatio-temporal data with complex dependency structures is a challenging task but can be simplified by assuming that the original variables are generated from independent latent components. If these components are found, they can be modelled univariately. Blind source separation aims to recover the latent components by estimating the unknown linear or nonlinear unmixing transformation based on the observed data only. In this paper, we extend recently introduced identifiable variational autoencoder to the nonlinear nonstationary spatio-temporal blind source separation setting and demonstrate its performance using comprehensive simulation studies. Additionally, we introduce two alternative methods for the latent dimension estimation, which is a crucial task in order to obtain the correct latent representation. Finally, we illustrate the proposed methods using a meteorological application, where we estimate the latent dimension and the latent components, interpret the components, and show how nonstationarity can be accounted and prediction accuracy can be improved by using the proposed nonlinear blind source separation method as a preprocessing method.

Monte Carlo inference for semiparametric Bayesian regression

Data transformations are essential for broad applicability of parametric regression models. However, for Bayesian analysis, joint inference of the transformation and model parameters typically involves restrictive parametric transformations or nonparametric representations that are computationally inefficient and cumbersome for implementation and theoretical analysis, which limits their usability in practice. This paper introduces a simple, general, and efficient strategy for joint posterior inference of an unknown transformation and all regression model parameters. The proposed approach directly targets the posterior distribution of the transformation by linking it with the marginal distributions of the independent and dependent variables, and then deploys a Bayesian nonparametric model via the Bayesian bootstrap. Crucially, this approach delivers (1) joint posterior consistency under general conditions, including multiple model misspecifications, and (2) efficient Monte Carlo (not Markov chain Monte Carlo) inference for the transformation and all parameters for important special cases. These tools apply across a variety of data domains, including real-valued, positive, and compactly-supported data. Simulation studies and an empirical application demonstrate the effectiveness and efficiency of this strategy for semiparametric Bayesian analysis with linear models, quantile regression, and Gaussian processes. The R package SeBR is available on CRAN.

Phosphonoalamides Reveal the Biosynthetic Origin of Phosphonoalanine Natural Products and a Convergent Pathway for Their Diversification

AbstractPhosphonate natural products, with their potent inhibitory activity, have found widespread use across multiple industries. Their success has inspired development of genome mining approaches that continue to reveal previously unknown bioactive scaffolds and biosynthetic insights. However, a greater understanding of phosphonate metabolism is required to enable prediction of compounds and their bioactivities from sequence information alone. Here, we expand our knowledge of this natural product class by reporting the complete biosynthesis of the phosphonoalamides, antimicrobial tripeptides with a conserved N‐terminal l‐phosphonoalanine (PnAla) residue produced by Streptomyces. The phosphonoalamides result from the convergence of PnAla biosynthesis and peptide ligation pathways. We elucidate the biochemistry underlying the transamination of phosphonopyruvate to PnAla, a new early branchpoint in phosphonate biosynthesis catalyzed by an aminotransferase with evolved specificity for phosphonate metabolism. Peptide formation is catalyzed by two ATP‐grasp ligases, the first of which produces dipeptides, and a second which ligates dipeptides to PnAla to produce phosphonoalamides. Substrate specificity profiling revealed a dramatic expansion of dipeptide and tripeptide products, while finding PnaC to be the most promiscuous dipeptide ligase reported thus far. Our findings highlight previously unknown transformations in natural product biosynthesis, promising enzyme biocatalysts, and unveil insights into the diversity of phosphonopeptide natural products.

Phosphonoalamides Reveal the Biosynthetic Origin of Phosphonoalanine Natural Products and a Convergent Pathway for Their Diversification.

Phosphonate natural products, with their potent inhibitory activity, have found widespread use across multiple industries. Their success has inspired development of genome mining approaches that continue to reveal previously unknown bioactive scaffolds and biosynthetic insights. However, a greater understanding of phosphonate metabolism is required to enable prediction of compounds and their bioactivities from sequence information alone. Here, we expand our knowledge of this natural product class by reporting the complete biosynthesis of the phosphonoalamides, antimicrobial tripeptides with a conserved N-terminal l-phosphonoalanine (PnAla) residue produced by Streptomyces. The phosphonoalamides result from the convergence of PnAla biosynthesis and peptide ligation pathways. We elucidate the biochemistry underlying the transamination of phosphonopyruvate to PnAla, a new early branchpoint in phosphonate biosynthesis catalyzed by an aminotransferase with evolved specificity for phosphonate metabolism. Peptide formation is catalyzed by two ATP-grasp ligases, the first of which produces dipeptides, and a second which ligates dipeptides to PnAla to produce phosphonoalamides. Substrate specificity profiling revealed a dramatic expansion of dipeptide and tripeptide products, while finding PnaC to be the most promiscuous dipeptide ligase reported thus far. Our findings highlight previously unknown transformations in natural product biosynthesis, promising enzyme biocatalysts, and unveil insights into the diversity of phosphonopeptide natural products.

LPOT: Locality-Preserving Gromov-Wasserstein Discrepancy for Nonrigid Point Set Registration.

The main problems in point registration involve recovering correspondences and estimating transformations, especially in a fully unsupervised way without any feature descriptors. In this work, we propose a robust point matching method using discrete optimal transport (OT), which is a natural and useful approach for assignment tasks, to recover the underlying correspondences and improve the nonrigid registration in the presence of unknown global transformations. Specifically, we cast the registration problem as a joint estimation over local transport couplings and global transformations, observing that the local neighborhood topology structures should be preserved strongly and stably for nonrigid transformations. By solving the Gromov-Wasserstein discrepancy, a smooth assignment matrix from one point set to another can be recovered in a fully unsupervised way. Registration performance can be improved by applying an unsupervised map to guide the transformation estimate under the alternating optimization. Experimental results on several datasets reveal how the presented method is superior to the state-of-the-art methods when facing large data degradations.

Efficient parameter estimation for ODE models of cellular processes using semi-quantitative data.

Quantitative dynamical models facilitate the understanding of biological processes and the prediction of their dynamics. The parameters of these models are commonly estimated from experimental data. Yet, experimental data generated from different techniques do not provide direct information about the state of the system but a nonlinear (monotonic) transformation of it. For such semi-quantitative data, when this transformation is unknown, it is not apparent how the model simulations and the experimental data can be compared. We propose a versatile spline-based approach for the integration of a broad spectrum of semi-quantitative data into parameter estimation. We derive analytical formulas for the gradients of the hierarchical objective function and show that this substantially increases the estimation efficiency. Subsequently, we demonstrate that the method allows for the reliable discovery of unknown measurement transformations. Furthermore, we show that this approach can significantly improve the parameter inference based on semi-quantitative data in comparison to available methods. Modelers can easily apply our method by using our implementation in the open-source Python Parameter EStimation TOolbox (pyPESTO) available at https://github.com/ICB-DCM/pyPESTO.

Review of an up-to-date, emphatic, and concise perspective on e-cigarettes

Electronic cigarettes entered the market in 2007 as an alternative to conventional tobacco products, and their use has increased significantly worldwide. Due to the emergence of various harmful effects and attracting attention, their sale was banned in our country and many other countries. However, according to the research, the use of e-cigarettes is increasing rapidly in the world and in Türkiye, and the rate of use at young ages is quite high especially. Its supply can be easily made through online sales on many websites. Despite this growing popularity, little is known about the potential effects of e-cigarettes on human health. This is particularly important in the presence of flavoring compounds, solvents, additives, and other ingredients that are added intentionally or unintentionally and whose long-term effects are uncertain. This is because every day a new product with new ingredients, flavors, and properties is introduced to the market. Previous studies have focused on measuring known toxicants, particularly those found in traditional cigarettes, while fewer have investigated unknown compounds and transformation products formed during the e-cigarette process in these diverse and ever-evolving products. It seems that the e-cigarette market is growing and developing rapidly in our country, as in the whole world, and product sales continue to increase despite all kinds of bans. However, when we look at the literature, the research conducted is insufficient, as mentioned above, and there are points that need to be further emphasized.

Reaction of Picolinamides with Ketones Producing a New Type of Heterocyclic Salts with an Imidazolidin-4-One Ring.

Reactions of picolinamides with 1,3-propanesultone in methanol followed by the treatment with ketones led to a series of previously unknown chemical transformations, yielding first pyridinium salts (2a-f), with a protonated endocyclic nitrogen atom, and then heterocyclic salts (3a-j) containing an imidazolidin-4-one ring. The structures of intermediate and final products were determined by IR and 1H, 13C NMR spectroscopy, and X-ray study. The effects of the ketone and alcohol structures on the product yield were studied by quantum-chemical calculations. The stability of salts 3a-j towards hydrolysis and alcoholysis makes them excellent candidates for the search for new types of biologically active compounds.

Screening and identification of unknown chemical contaminants in food based on liquid chromatography–high-resolution mass spectrometry and machine learning

Unknown or unexpected chemical contaminants and/or their transformation products in food that may be harmful to humans need to be discovered for comprehensive safety evaluation. Liquid chromatography–high-resolution mass spectrometry (LC-HRMS) is a powerful tool for detecting chemical contaminants in food samples. However, identifying all of peaks in LC-HRMS is not possible, but if class information is known in advance, further identification will become easier. In this work, a novel MS2 spectra classification-driven screening strategy was constructed based on LC-HRMS and machine learning. First, the classification model was developed based on machine learning algorithm using class information and experimental MS2 data of chemical contaminants and other non-contaminants. By using the developed artificial neural network classification model, in total 32 classes of pesticides, veterinary drugs and mycotoxins were classified with good prediction accuracy and low false-positive rate. Based on the classification model, a screening procedure was developed in which the classes of unknown features in LC-HRMS were first predicted through the classification model, and then their structures were identified under the guidance of class information. Finally, the developed strategy was tentatively applied to the analysis of pork and aquatic products, and 8 chemical contaminants and 11 transformation products belonging to 8 classes were found. This strategy enables screening of unknown chemical contaminants and transformation products in complex food matrices.

Uptake and metabolization of four sartan drugs by eight different plants: Targeted and untargeted analyses by HPLC-drift-tube-ion-mobility quadrupole time-of-flight mass spectrometry.

In this study, we investigated the uptake and metabolization of four drugs (plus the associated prodrugs) from the sartan family by eight edible plants. Growing the plants hydroponically in a medium containing the respective drug, more than 40 phases I and II metabolites derived from the four sartan drugs could be tentatively identified. To demonstrate the suitability of the proposed analytical approach for actual environmental samples, garden cress (Lepidium sativum) selected as a model plant was grown in water drawn from the effluent of two local wastewater treatment plants. Thereby, three of the sartans, namely, olmesartan, candesartan, and valsartan, could be found in the plant extracts at concentrations of 3.1, 10.4, and 14.4ngg-1 , respectively. Additionally, for candesartan and valsartan, a glycosylated transformation product could be detected. In order to extend the present (targeted) workflow also toward the analysis of unknown transformation products (i.e., those not listed in the custom-made database used for this research), a nontargeted approach for the analysis of plant extracts with respect to the presence of drug-related metabolites was developed. Comparison of the targeted and the nontargeted workflows led to the finding of two additional, so far unidentified, transformation products originating from azilsartan.

Photo-assisted transformation of furosemide: Exploring transformation pathways, structure database and suspect and non-target workflows for comprehensive screening of unknown transformation products in wastewaters and landfill leachates

In recent years, transformation products-(TPs) of pharmaceuticals in the environment have received considerable attention. In this context, here, a customized overview of transformation of Furosemide-(FRS) in aqueous matrices treated by photo-oxidation is provided as a proof of concept. Hence, the primary goal of the study was to display an integrated strategy by combining the target (parent-molecule) and suspect screening-(SS) approaches (TPs) in order to build an in-house High-Resolution mass spectrometry (HRMS) database able to provide reference information (chromatographic/spectral) for environmental investigations in complex matrices (wastewaters/landfill leachates). Data analysis was performed by optimizing a SS workflow. Additional confirmation for the proposed structural elucidation was provided by correlating retention time to the proposed structure employing three prediction models. This approach was applied for the tentative identification of 35 TPs of FRS, 28 of which are reported herein for the first time. Finally, SS and non-target analysis (NTA) have been successfully applied for retrospective screening of FRS and its TPs in real samples. The findings demonstrated that SS allows the proper identification of TPs of FRS in complex matrices proving its outstanding importance compared to NTA. In total, six TPs were identified by SS with potential ecotoxicological implications for two of them according to in silico risk assessment.

Electrochemistry-mass spectrometry bridging the gap between suspect and target screening of valsartan transformation products in wastewater treatment plant effluent

Degradation of xenobiotics in wastewater treatment plants may lead to the formation of transformation products with higher persistence or increased (eco-)toxic potential compared to the parent compounds. Accordingly, the identification of transformation products from wastewater treatment plant effluents has gained increasing attention. Here, we show the potential of electrochemistry hyphenated to liquid chromatography and mass spectrometry for the prediction of oxidative degradation in wastewater treatment plants using the antihypertensive drug valsartan as a model compound. This approach identifies seven electrochemical transformation products of valsartan, which are used to conduct a suspect screening in effluent of the main wastewater treatment plant in the city of Münster in Germany. Apart from the parent compound valsartan, an electrochemically predicted transformation product, the N-dealkylated ETP336, is detected in wastewater treatment plant effluent. Subsequently, a targeted liquid chromatographytandem mass spectrometry method for the detection of valsartan and its electrochemical transformation products is set up. Here, electrochemical oxidation is used to generate reference materials of the transformation products in situ by hyphenating electrochemistry online to a triple quadrupole mass spectrometer. Using this setup, multiple reaction monitoring transitions are set up without the need for laborious and costly synthesis and isolation of reference materials for the transformation products. The targeted method is then applied to extracts from wastewater treatment plant effluent and the presence of ETP336 and valsartan in the samples is verified. The presented workflow can be used to set up targeted analysis methods for previously unknown transformation products even without the need for expensive high-resolution mass spectrometers.

Robust inference for subgroup analysis with general transformation models

A crucial step in developing personalized strategies in precision medicine or precision marketing is to identify the latent subgroups of patients or customers of a heterogeneous population. In this article, we consider a general class of heterogeneous transformation models for subgroup identification, under which an unknown monotonic transformation of the response is linearly related to the covariates via subject-specific regression coefficients with unknown error distribution. This class of models is broad enough to cover many popular models, including the heterogeneous linear model, the heterogeneous Cox’s proportional hazard model and the heterogeneous proportional odds model. Without any priori grouping information, we propose a robust method based on the maximum rank correlation and a concave fusion to automatically determine the number of subgroups, identify the latent subgroup structure, and estimate the subgroup-specific covariate effects simultaneously. We establish the theoretical properties of our proposed estimate under regularity conditions. A random weighting resampling scheme is used for variance estimation. The proposed procedure can be easily extended to handle censored data. Numerical studies including simulations and two real data analysis demonstrate that the proposed method performs reasonably well in practical situations.

Spearman Rank Correlation Screening for Ultrahigh-Dimensional Censored Data

Herein, we propose a Spearman rank correlation-based screening procedure for ultrahigh-dimensional data with censored response cases. The proposed method is model-free without specifying any regression forms of predictors or response variables and is robust under the unknown monotone transformations of these response variable and predictors. The sure-screening and rank-consistency properties are established under some mild regularity conditions. Simulation studies demonstrate that the new screening method performs well in the presence of a heavy-tailed distribution, strongly dependent predictors or outliers, and offers superior performance over the existing nonparametric screening procedures. In particular, the new screening method still works well when a response variable is observed under a high censoring rate. An illustrative example is provided.

Learning Maximum Margin Channel Decoders

The problem of learning a channel decoder is considered for two channel models. The first model is an additive noise channel whose noise distribution is unknown and nonparametric. The learner is provided with a fixed codebook and a dataset comprised of independent samples of the noise, and is required to select a precision matrix for a nearest neighbor decoder in terms of the Mahalanobis distance. The second model is a non-linear channel with additive white Gaussian noise and unknown channel transformation. The learner is provided with a fixed codebook and a dataset comprised of independent input-output samples of the channel, and is required to select a matrix for a nearest neighbor decoder with a linear kernel. For both models, the objective of maximizing the margin of the decoder is addressed. Accordingly, for each channel model, a regularized loss minimization problem with a codebook-related regularization term and hinge-like loss function is developed, which is inspired by the support vector machine paradigm for classification problems. Expected generalization error bounds for the error probability loss function are provided for both models, under optimal choice of the regularization parameter. For the additive noise channel, a theoretical guidance for choosing the training signal-to-noise ratio is proposed based on this bound. In addition, for the non-linear channel, a high probability uniform generalization error bound is provided for the hypothesis class. For each channel, a stochastic sub-gradient descent algorithm for solving the regularized loss minimization problem is proposed, and an optimization error bound is stated. The performance of the proposed algorithms is demonstrated through several examples.

Non-targeted screening of emerging contaminants in South African surface and wastewater

Targeted analysis of contaminants in water are often focused on a narrow range of chemicals, falling short of the true pollution status of water bodies. Non targeted screening presents several advantages in identifying less prioritized pollutants, entirely unknown compounds and transformation products. In this study we employed a non-targeted screening workflow established on a high-resolution quadrupole time-of-flight mass spectrometer coupled to an ultrahigh performance liquid chromatograph (UHPLC-QTOF-MS) to identify known and unknown pollutants in South African waste and surface waters. Level two confidence identification of 315 compounds was achieved based on mass accuracy, isotope patterns and MS/MS spectra match. Pharmaceuticals, drugs, and metabolites made up 40% of the detected compounds, biological compounds and industrial chemicals along with their metabolites constituted 24 and 18% respectively, while personal care products, pesticides and food additives made up approximately 5, 4 and 4% respectively. Several antiretroviral drugs were confirmed with level one confidence using isotope labelled standards. A wide range of “new” pharmaceuticals, pesticides and metabolites were documented in South African waters for the first time. Seventeen (17) pharmaceuticals were reported for the first time in South African waters, of which four are reported for the first time ever in surface water.The result of this exploratory study highlights the presence of several contaminants of public health concern that have hitherto received little to no attention in previous wastewater-based epidemiological studies. We provide a detailed list of priority contaminants for future studies on targeted-analysis.

Identification, occurrence, concentration and composition profile of fiproles in municipal wastewater treatment plants

Although fipronil and several of its transformation products are ubiquitous in aquatic environments, limited information is available on the structural identities, detection frequencies, concentrations and composition profiles of fiproles (fipronil and its known and unknown transformation products) in municipal wastewater treatment plants (WWTPs). In this study, a suspect screening analysis was applied to identify and characterize fipronil transformation products in 16 municipal WWTPs from three cities in China. In addition to fipronil and its four transformation products (fipronil amide, fipronil sulfide, fipronil sulfone and desulfinyl fipronil), fipronil chloramine and fipronil sulfone chloramine were detected for the first time in municipal wastewater. Moreover, the cumulative concentrations of six transformation products were 0.236 ng/L and 3.44 ng/L in wastewater influents and effluents, and accounted for one-third (in influents) to half (in effluents) of fiproles. Of those transformation products, two chlorinated byproducts (fipronil chloramine and fipronil sulfone chloramine) were major transformation products in both municipal wastewater influents and effluents. Notably, the log Kow and bioconcentration factor (evaluated by EPI Suite software) of fipronil chloramine (log Kow = 6.64, and BCF = 11,200 L/kg wet-wt) and fipronil sulfone chloramine (log Kow = 4.42, and BCF = 382.9 L/kg wet-wt) were greater than that of their parent compound. Considering the persistence, bioaccumulation potential and toxicity, the high detection rates of fipronil chloramine and fipronil sulfone chloramine in urban aquatic systems need to be specifically considered in future ecological risk assessments.

Intelligent Workflow and Software for Non-Target Analysis of Complex Samples Using a Mixture of Toxic Transformation Products of Unsymmetrical Dimethylhydrazine as an Example.

Unsymmetrical dimethylhydrazine (UDMH) is a widely used rocket propellant. Entering the environment or being stored in uncontrolled conditions, UDMH easily forms an enormous variety (at least many dozens) of transformation products. Environmental pollution by UDMH and its transformation products is a major problem in many countries and across the Arctic region. Unfortunately, previous works often use only electron ionization mass spectrometry with a library search, or they consider only the molecular formula to propose the structures of new products. This is quite an unreliable approach. It was demonstrated that a newly proposed artificial intelligence-based workflow allows for the proposal of structures of UDMH transformation products with a greater degree of certainty. The presented free and open-source software with a convenient graphical user interface facilitates the non-target analysis of industrial samples. It has bundled machine learning models for the prediction of retention indices and mass spectra. A critical analysis of whether a combination of several methods of chromatography and mass spectrometry allows us to elucidate the structure of an unknown UDMH transformation product was provided. It was demonstrated that the use of gas chromatographic retention indices for two stationary phases (polar and non-polar) allows for the rejection of false candidates in many cases when only one retention index is not enough. The structures of five previously unknown UDMH transformation products were proposed, and four previously proposed structures were refined.

Abstract P3-08-04: Discovery of cancer genes and pathways operative in PI3K-activated mammary cancer reveals clinically relevant genotype-phenotype correlations

Abstract Human breast cancer (BRCA) shows tremendous genomic, gene expression, clinical, and phenotypic heterogeneity. Known driver gene alterations can only explain a portion of this heterogeneity, some of which likely arise from variation in the target cell for transformation, in addition to incompletely understood gene copy number and epigenetic alterations. These factors are difficult to identify with certainty using human patient samples due to widely varying germline genetic backgrounds, thousands of gene copy and epigenetic changes per sample, and, unknown target cell transformation. Activating mutations in the p110α catalytic subunit of PI3K are one of the most common genetic alterations in human BRCA. Here, we report results from two Sleeping Beauty (SB) transposon-accelerated mouse models of Pik3ca-mutant mammary cancer showing how genotype-phenotype correlations can be drawn providing strong candidates for mediating tumor phenotypes, including estrogen-receptor (ER)-dependent gene expression, high cell cycle activity, and immune cell exclusion. We used SB transposon mutagenesis in mice on a Pik3caH1047R activated mutant background to model mammary cancer development in two different mammary epithelial compartments. Both the target cell for mutagenesis and the specific transposon-induced mutations correlated with specific tumor phenotypes, including whether the tumors were ER positive or negative. RNA sequencing of tumors revealed novel genotype-phenotype correlations implicating specific transposon-altered gene drivers of high cell cycle activity, ER-dependent gene expression, and white blood cell exclusion from the tumor. Many transposon-implicated genes are altered at the gene copy number or epigenetic/methylation level in human BRCA, and several were functionally validated. These models provide a source of genetically heterogenous mouse mammary tumors with a uniform initiating mutation, Pik3caH1047R, useful for identifying cooperating pathways and drivers of specific tumor phenotypes. Citation Format: Morito Kurata, Emiily Pope, Jingmin Shu, Wenlin Yuan, Wendy Hudson, Mark Sokolowski, Setareh Bagherzadeh, Zora Modrusan, Eric Stawiski, Steffen Durinck, Sekar Seshigiri, Aaron Sarver, Nuri Temiz, David Largaespada. Discovery of cancer genes and pathways operative in PI3K-activated mammary cancer reveals clinically relevant genotype-phenotype correlations. [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P3-08-04.