TDC-2: Multimodal Foundation for Therapeutic Science.

Therapeutics Data Commons (tdcommons.ai) is an open science initiative with unified datasets, AI models, and benchmarks to support research across therapeutic modalities and drug discovery and development stages. The Commons 2.0 (TDC-2) is a comprehensive overhaul of Therapeutic Data Commons to catalyze research in multimodal models for drug discovery by unifying single-cell biology of diseases, biochemistry of molecules, and effects of drugs through multimodal datasets, AI-powered API endpoints, new multimodal tasks and model frameworks, and comprehensive benchmarks. TDC-2 introduces over 1,000 multimodal datasets spanning approximately 85 million cells, pre-calculated embeddings from 5 state-of-the-art single-cell models, and a biomedical knowledge graph. TDC-2 drastically expands the coverage of ML tasks across therapeutic pipelines and 10+ new modalities, spanning but not limited to single-cell gene expression data, clinical trial data, peptide sequence data, peptidomimetics protein-peptide interaction data regarding newly discovered ligands derived from AS-MS spectroscopy, novel 3D structural data for proteins, and cell-type-specific protein-protein interaction networks at single-cell resolution. TDC-2 introduces multimodal data access under an API-first design using the model-view-controller paradigm. TDC-2 introduces 7 novel ML tasks with fine-grained biological contexts: contextualized drug-target identification, single-cell chemical/genetic perturbation response prediction, protein-peptide binding affinity prediction task, and clinical trial outcome prediction task, which introduce antigen-processing-pathway-specific, cell-type-specific, peptide-specific, and patient-specific biological contexts. TDC-2 also releases benchmarks evaluating 15+ state-of-the-art models across 5+ new learning tasks evaluating models on diverse biological contexts and sampling approaches. Among these, TDC-2 provides the first benchmark for context-specific learning. TDC-2, to our knowledge, is also the first to introduce a protein-peptide binding interaction benchmark.

Site-Specific Antibody Prodrugs via S-Arylation: a Bioconjugation Approach Toward Masked Tyrosine Analogues.

The utility of antibody therapeutics is hampered by potential cross-reactivity with healthy tissue. Over the past decade, significant advances have been made in the design of activatable antibodies, which increase, or create altogether, the therapeutic window of a parent antibody. Of these, antibody prodrugs (pro-antibodies) are masked antibodies that have advanced the most for therapeutic use. They are designed to reveal the active, parent antibody only when encountering proteases upregulated in the microenvironment of the targeted disease tissue, thereby minimizing off-target activity. However, current pro-antibody designs are relegated to fusion proteins that append masking groups restricted to the use of only canonical amino acids, offering excellent control of the site of introduction, but with no authority over where the masking group is installed other than the N-terminus of the antibody. Here, we present a palladium-based bioconjugation approach for the site-specific introduction of a masked tyrosine mimic in the complementary determining region of the FDA approved antibody therapeutic ipilimumab used as a model system. The approach enables the introduction of a protease cleavable group tethered to noncanonical polymers (polyethylene glycol (PEG)) resulting in 47-fold weaker binding to cells expressing CTLA-4, the target antigen of ipilimumab. Upon exposure to tumor-associated proteases, the masking group is cleaved, unveiling a tyrosine-mimic (dubbed hydroxyphenyl cysteine (HPC)) that restores (>90% restoration) binding affinity to its target antigen.

Early Implementation and Outcomes Among People with HIV Who Accessed Long-Acting Injectable Cabotegravir/Rilpivirine at Two Ryan White Clinics in the U.S. South.

The use of long-acting injectable cabotegravir/rilpivirine (LAI-CAB/RPV) as maintenance therapy for persons with HIV (PWH), which may improve treatment access and outcomes, though real-world data on uptake are limited, was studied at two Ryan White clinics in Atlanta, Georgia. Among PWH referred from 4/1/2021 to 9/15/2022 to switch to LAI-CAB/RPV, characteristics were ascertained at time of referral; and disposition (initiated; ineligible; uninterested; pending) was recorded as of 9/15/2022. Among patients initiated on CAB/RPV, we assessed the drug procurement process and clinical outcomes through 6/1/2023. Among 149 PWH referred, 74/149 (50%) initiated CAB/RPV as of 9/15/2022, of whom, characteristics were median age 47 (Q1-Q3 36-55) years, 16% cisgender female, 72% Black race, median HIV duration 15 (Q1-Q3 9-19) years, and 64% had commercial health insurance. Of the 75 PWH not initiated, 35 were ineligible owing to a clinical concern (n = 16) or insurance issue (n = 19); 15 patients changed their mind about switching; and 25 were pending eligibility review or therapy initiation. Median time from CAB/RPV prescription to initiation was 46 (Q1-Q3 29-78) days. Of 731 total injections administered (median 11 injections/patient), 95% were given within 7 days of the target treatment date. Nearly all patients were virally suppressed upon referral and remained suppressed through follow-up. At two clinics in the U.S. South, half of the patients referred for LAI-CAB/RPV successfully accessed therapy nearly 2 years after U.S. drug approval. We identified barriers to uptake at the patient and structural levels, highlighting key areas to invest resource and personnel support to sustain and scale long-acting antiretroviral therapy programming.

Watershed analysis of urban stormwater contaminant 6PPD-Quinone hotspots and stream concentrations using a process-based ecohydrological model.

Coho salmon (Oncorhynchus kisutch) are highly sensitive to 6PPD-Quinone (6PPD-Q). Details of the hydrological and biogeochemical processes controlling spatial and temporal dynamics of 6PPD-Q fate and transport from points of deposition to receiving waters (e.g., streams, estuaries) are poorly understood. To understand the fate and transport of 6PPD and mechanisms leading to salmon mortality Visualizing Ecosystem Land Management Assessments (VELMA), an ecohydrological model developed by US Environmental Protection Agency (EPA), was enhanced to better understand and inform stormwater management planning by municipal, state, and federal partners seeking to reduce stormwater contaminant loads in urban streams draining to the Puget Sound National Estuary. This work focuses on the 5.5 km2 Longfellow Creek upper watershed (Seattle, Washington, United States), which has long exhibited high rates of acute urban runoff mortality syndrome in coho salmon. We present VELMA model results to elucidate these processes for the Longfellow Creek watershed across multiple scales-from 5-m grid cells to the entire watershed. Our results highlight hydrological and biogeochemical controls on 6PPD-Q flow paths, and hotspots within the watershed and its stormwater infrastructure, that ultimately impact contaminant transport to Longfellow Creek and Puget Sound. Simulated daily average 6PPD-Q and available observed 6PPD-Q peak in-stream grab sample concentrations (ng/L) corresponds within plus or minus 10 ng/L. Most importantly, VELMA's high-resolution spatial and temporal analysis of 6PPD-Q hotspots provides a tool for prioritizing the locations, amounts, and types of green infrastructure that can most effectively reduce 6PPD-Q stream concentrations to levels protective of coho salmon and other aquatic species.

We've all been wrong about provisional tic disorder.

BACKGROUND: Provisional Tic Disorder (PTD) is common in childhood. The received wisdom among clinicians is that PTD is short-lived and mild, with at most a few tics, and rarely includes complex tics, premonitory phenomena or comorbid illnesses. However, such conclusions come from clinical experience, with biased ascertainment and limited follow-up. METHODS: Prospective study of 89 children with tics starting 0-9 months ago (median 4 months), fewer than half from clinical sources. Follow-up at 12 (± 24, 36, 48) months after the first tic. RESULTS: At study entry, many children had ADHD (39), an anxiety disorder (27), OCD (9) or enuresis (17). All had at least two current tics, with a mean total since onset of 6.9 motor and 2.0 phonic tics. Forty-one had experienced a complex tic, and 69 could suppress some tics. Tics were clinically meaningful: 64 had tics severe enough for a clinical trial, and 76 families sought medical attention for the tics. At 12 months, 79 returned, and 78 still had tics. Of these, 29 manifested no tics during history and extended examination, but only via audio-visual monitoring when the child was seated alone. Only 12/70 now had plans to see a doctor for tics. Most who returned at 2-4 years still had tics known to the child and family, but medical impact was low. CONCLUSIONS: Our results do not contradict previous data, but overturn clinical lore. The data strongly argue against the longstanding but arbitrary tradition of separating tic disorders into recent-onset versus chronic.

Mapping brain function in adults and young children during naturalistic viewing with high-density diffuse optical tomography.

Human studies of early brain development have been limited by extant neuroimaging methods. MRI scanners present logistical challenges for imaging young children, while alternative modalities like functional near-infrared spectroscopy have traditionally been limited by image quality due to sparse sampling. In addition, conventional tasks for brain mapping elicit low task engagement, high head motion, and considerable participant attrition in pediatric populations. As a result, typical and atypical developmental trajectories of processes such as language acquisition remain understudied during sensitive periods over the first years of life. We evaluate high-density diffuse optical tomography (HD-DOT) imaging combined with movie stimuli for high resolution optical neuroimaging in awake children ranging from 1 to 7 years of age. We built an HD-DOT system with design features geared towards enhancing both image quality and child comfort. Furthermore, we characterized a library of animated movie clips as a stimulus set for brain mapping and we optimized associated data analysis pipelines. Together, these tools could map cortical responses to movies and contained features such as speech in both adults and awake young children. This study lays the groundwork for future research to investigate response variability in larger pediatric samples and atypical trajectories of early brain development in clinical populations.

Reversibly Reactive Affinity Selection-Mass Spectrometry Enables Identification of Covalent Peptide Binders.

Covalent peptide binders have found applications as activity-based probes and as irreversible therapeutic inhibitors. Currently, there is no rapid, label-free, and tunable affinity selection platform to enrich covalent reactive peptide binders from synthetic libraries. We address this challenge by developing a reversibly reactive affinity selection platform termed ReAct-ASMS enabled by tandem high-resolution mass spectrometry (MS/MS) to identify covalent peptide binders to native protein targets. It uses mixed disulfide-containing peptides to build reversible peptide-protein conjugates that can enrich for covalent variants, which can be sequenced by MS/MS after reduction. Using this platform, we identified covalent peptide binders against two oncoproteins, human papillomavirus 16 early protein 6 (HPV16 E6) and peptidyl-prolyl cis-trans isomerase NIMA-interacting 1 protein (Pin1). The resulting peptide binders efficiently and selectively cross-link Cys58 of E6 at 37 °C and Cys113 of Pin1 at room temperature, respectively. ReAct-ASMS enables the identification of highly selective covalent peptide binders for diverse molecular targets, introducing an applicable platform to assist preclinical therapeutic development pipelines.

Assessing the Performance of Peptide Force Fields for Modeling the Solution Structural Ensembles of Cyclic Peptides.

Molecular dynamics simulation is a powerful tool for characterizing the solution structural ensembles of cyclic peptides. However, the ability of simulation to recapitulate experimental results and make accurate predictions largely depends on the force fields used. In our work here, we evaluate the performance of seven state-of-the-art force fields in recapitulating the experimental NMR results in water of 12 benchmark cyclic peptides, consisting of 6 cyclic pentapeptides, 4 cyclic hexapeptides, and 2 cyclic heptapeptides. The results show that RSFF2+TIP3P, RSFF2C+TIP3P, and Amber14SB+TIP3P exhibit similar and the best performance, all recapitulating the NMR-derived structure information on 10 cyclic peptides. Amber19SB+OPC successfully recapitulates the NMR-derived structure information on 8 cyclic peptides. In contrast, OPLS-AA/M+TIP4P, Amber03+TIP3P, and Amber14SBonlysc+GB-neck2 could only recapitulate the NMR-derived structure information on 5 cyclic peptides, the majority of which are not well-structured.

Long-Circulating Vasoactive 1,18-Octadecanedioic Acid-Terlipressin Conjugate.

Hepatorenal syndrome (HRS) is a life-threatening complication of end-stage liver disease first reported over a century ago, but its management still poses an unmet challenge. A therapeutic agent found to stabilize the condition is a short cyclic peptide, vasopressin analogue, terlipressin (TP). While TP is commonly prescribed for HRS patients in most parts of the world, it was only recently approved for use in the United States. TP exhibits short circulation half-lives and adverse side effects associated with the dose required. Herein, we present a 1,18-octadecanedioic acid (ODDA) conjugate of the cyclic peptide (ODDA-TP), which enables noncovalent binding to serum albumin via native fatty acid binding modes. ODDA-TP is demonstrated to outperform TP alone in studies including in vitro cellular receptor activation, stability in plasma, pharmacokinetics, and performance in vivo in rats. Specifically, ODDA-TP had an elimination half-life 20 times that of TP alone while exhibiting a superior safety profile.

Development of a synthetic relaxin-3/INSL5 chimeric peptide ligand for NanoBiT complementation binding assays.

INSL5 and relaxin-3 are relaxin family peptides with important roles in gut and brain function, respectively. They mediate their actions through the class A GPCRs RXFP4 and RXFP3. RXFP4 has been proposed to be a therapeutic target for colon motility disorders whereas RXFP3 targeting could be effective for neurological conditions such as anxiety. Validation of these targets has been limited by the lack of specific ligands and the availability of robust ligand-binding assays for their development. In this study, we have utilized NanoBiT complementation to develop a SmBiT-conjugated tracer for use with LgBiT-fused RXFP3 and RXFP4. The low affinity between LgBiT:SmBiT should result in a low non-specific luminescence signal and enable the quantification of binding without the tedious separation of non-bound ligands. We used solid-phase peptide synthesis to produce a SmBiT-labelled RXFP3/4 agonist, R3/I5, where SmBiT was conjugated to the B-chain N-terminus via a PEG12 linker. Both SmBiT-R3/I5 and R3/I5 were synthesized and purified in high purity and yield. Stable HEK293T cell lines expressing LgBiT-RXFP3 and LgBiT-RXFP4 were produced and demonstrated normal signaling in response to the synthetic R3/I5 peptide. Binding was first characterized in whole-cell binding kinetic assays validating that the SmBiT-R3/I5 bound to both cell lines with nanomolar affinity with minimal non-specific binding without bound and free SmBiT-R3/I5 separation. We then optimized membrane binding assays, demonstrating easy and robust analysis of both saturation and competition binding from frozen membranes. These assays therefore provide an appropriate rigorous binding assay for the high-throughput analysis of RXFP3 and RXFP4 ligands.

How hydrophobicity, side chains, and salt affect the dimensions of disordered proteins.

Despite the generally accepted role of the hydrophobic effect as the driving force for folding, many intrinsically disordered proteins (IDPs), including those with hydrophobic content typical of foldable proteins, behave nearly as self-avoiding random walks (SARWs) under physiological conditions. Here, we tested how temperature and ionic conditions influence the dimensions of the N-terminal domain of pertactin (PNt), an IDP with an amino acid composition typical of folded proteins. While PNt contracts somewhat with temperature, it nevertheless remains expanded over 10-58°C, with a Flory exponent, ν, >0.50. Both low and high ionic strength also produce contraction in PNt, but this contraction is mitigated by reducing charge segregation. With 46% glycine and low hydrophobicity, the reduced form of snow flea anti-freeze protein (red-sfAFP) is unaffected by temperature and ionic strength and persists as a near-SARW, ν ~ 0.54, arguing that the thermal contraction of PNt is due to stronger interactions between hydrophobic side chains. Additionally, red-sfAFP is a proxy for the polypeptide backbone, which has been thought to collapse in water. Increasing the glycine segregation in red-sfAFP had minimal effect on ν. Water remained a good solvent even with 21 consecutive glycine residues (ν > 0.5), and red-sfAFP variants lacked stable backbone hydrogen bonds according to hydrogen exchange. Similarly, changing glycine segregation has little impact on ν in other glycine-rich proteins. These findings underscore the generality that many disordered states can be expanded and unstructured, and that the hydrophobic effect alone is insufficient to drive significant chain collapse for typical protein sequences.

Automated Flow Peptide Synthesis Enables Engineering of Proteins with Stabilized Transient Binding Pockets.

Engineering at the amino acid level is key to enhancing the properties of existing proteins in a desired manner. So far, protein engineering has been dominated by genetic approaches, which have been extremely powerful but only allow for minimal variations beyond the canonical amino acids. Chemical peptide synthesis allows the unrestricted incorporation of a vast set of unnatural amino acids with much broader functionalities, including the incorporation of post-translational modifications or labels. Here we demonstrate the potential of chemical synthesis to generate proteins in a specific conformation, which would have been unattainable by recombinant protein expression. We use recently established rapid automated flow peptide synthesis combined with solid-phase late-stage modifications to rapidly generate a set of FK506-binding protein 51 constructs bearing defined intramolecular lactam bridges. This trapped an otherwise rarely populated transient pocket-as confirmed by crystal structures-which led to an up to 39-fold improved binding affinity for conformation-selective ligands and represents a unique system for the development of ligands for this rare conformation. Overall, our results show how rapid automated flow peptide synthesis can be applied to precision protein engineering.

Developing insulin-like peptide 5-based antagonists for the G protein-coupled receptor, RXFP4.

Human insulin-like peptide 5 (INSL5) is a gut hormone produced by colonic L-cells, and its biological functions are mediated by Relaxin Family Peptide Receptor 4 (RXFP4). Our preliminary data indicated that RXFP4 agonists are potential drug leads for the treatment of constipation. More recently, we designed and developed a novel RXFP4 antagonist, A13-nR that was shown to block agonist-induced activity in cells and animal models. We showed that A13-nR was able to block agonist-induced increases in colon motility in mice of both genders that express the receptor, RXFP4. Our data also showed that colorectal propulsion induced by intracolonic administration of short-chain fatty acids was antagonized by A13-nR. Therefore, A13-nR is an important research tool and potential drug lead for the treatment of colon motility disorders, such as bacterial diarrhea. However, A13-nR acted as a partial agonist at high concentrations in vitro and demonstrated modest antagonist potency (∼35 nM). Consequently, the primary objective of this study is to pinpoint novel modifications to A13-nR that eliminate partial agonist effects while preserving or augmenting antagonist potency. In this work, we detail the creation of a series of A13-nR-modified analogues, among which analogues 3, 4, and 6 demonstrated significantly improved RXFP4 affinity (∼3 nM) with reduced partial agonist activity, enhanced antagonist potency (∼10 nM) and maximum agonist inhibition (∼80 %) when compared with A13-nR. These compounds have potential as candidates for further preclinical evaluations, marking a significant stride toward innovative therapeutics for colon motility disorders.

Design of Cytotoxic T Cell Epitopes by Machine Learning of Human Degrons.

Antigen processing is critical for therapeutic vaccines to generate epitopes for priming cytotoxic T cell responses against cancer and pathogens, but insufficient processing often limits the quantity of epitopes released. We address this challenge using machine learning to ascribe a proteasomal degradation score to epitope sequences. Epitopes with varying scores were translocated into cells using nontoxic anthrax proteins. Epitopes with a low score show pronounced immunogenicity due to antigen processing, but epitopes with a high score show limited immunogenicity. This work sheds light on the sequence-activity relationships between proteasomal degradation and epitope immunogenicity. We anticipate that future efforts to incorporate proteasomal degradation signals into vaccine designs will lead to enhanced cytotoxic T cell priming by these vaccines in clinical settings.

Development of an α-Klotho Recognizing High-Affinity Peptide Probe from In-Solution Enrichment.

The kidney, parathyroid gland, and choroid plexus express the aging-related transmembrane protein α-Klotho, a coreceptor of the fibroblast growth factor 23 (FGF23) receptor complex. Reduced α-Klotho levels are correlated with chronic kidney disease and other age-related diseases, wherein they are released from membranes into circulation. Klotho's potential physiological action as a hormone is of current scientific interest. Part of the challenges associated with advancing these studies, however, has been the long-standing difficulty in detecting soluble α-Klotho in biofluids. Here, we describe the discovery of peptides that recognize α-Klotho with high affinity and selectivity by applying in-solution size-exclusion-based affinity selection-mass spectrometry (AS-MS). After two rounds of AS-MS and subsequent N-terminal modifications, the peptides improved their binding affinity to α-Klotho by approximately 2300-fold compared to the reported starting peptide Pep-10, previously designed based on the C-terminal region of FGF23. The lead peptide binders were shown to enrich α-Klotho from cell lysates and to label α-Klotho in kidney cells. Our results further support the utility of in-solution, label-free AS-MS protocols to discover peptide-based binders to target proteins of interest with high affinity and selectivity, resulting in functional probes for biological studies.

Novel multiparametric bulk and single EV pipeline for adipose cell-specific biomarker discovery in paired human biospecimens.

Obesity is a global health crisis that contributes to morbidity and mortality worldwide. Obesity's comorbid association with a variety of diseases, from metabolic syndrome to neurodegenerative disease, underscores the critical need to better understand the pathobiology of obesity. Adipose tissue, once seen as an inert storage depot, is now recognized as an active endocrine organ, regulating metabolic and systemic homeostasis. Recent studies spotlight the theranostic utility of extracellular vesicles (EVs) as novel biomarkers and drivers of disease, including obesity-related complications. Adipose-derived EVs (ADEVs) have garnered increased interest for their roles in diverse diseases, however robust isolation and characterization protocols for human, cell-specific EV subsets are limited. Herein, we directly address this technical challenge by establishing a multiparametric analysis framework that leverages bulk and single EV characterization, mRNA phenotyping and proteomics of human ADEVs directly from paired visceral adipose tissue, cultured mature adipocyte conditioned media, and plasma from obese subjects undergoing bariatric surgery. Importantly, rigorous EV phenotyping at the tissue and cell-specific level identified top 'adipose liquid biopsy' candidates that were validated in circulating plasma EVs from the same patient. In summary, our study paves the way toward a tissue and cell-specific, multiparametric framework for studying tissue and circulating adipose EVs in obesity-driven disease.

An executive-functions-based reading training enhances sensory-motor systems integration during reading fluency in children with dyslexia.

The Simple View of Reading model suggests that intact language processing and word decoding lead to proficient reading comprehension, with recent studies pointing at executive functions as an important component contributing to reading proficiency. Here, we aimed to determine the underlying mechanism(s) for these changes. Participants include 120 8- to 12-year-old children (n = 55 with dyslexia, n = 65 typical readers) trained on an executive functions-based reading program, including pre/postfunctional MRI and behavioral data collection. Across groups, improved word reading was related to stronger functional connections within executive functions and sensory networks. In children with dyslexia, faster and more accurate word reading was related to stronger functional connections within and between sensory networks. These results suggest greater synchronization of brain systems after the intervention, consistent with the "neural noise" hypothesis in children with dyslexia and support the consideration of including executive functions as part of the Simple View of Reading model.

When simplicity triumphs: niche specialization of gut bacteria exists even for simple fiber structures.

Structurally complex corn bran arabinoxylan (CAX) was used as a model glycan to investigate gut bacteria growth and competition on different AX-based fine structures. Nine hydrolyzate segments of the CAX polymer varying in chemical structure (sugars and linkages), CAX, five less complex non-corn arabinoxylans, and xylose and glucose were ranked from structurally complex to simple. The substrate panel promoted different overall growth and rates of growth of eight Bacteroides xylan-degrading strains. For example, Bacteroides cellulosilyticus DSM 14838 (Bacteroides cellulosilyticus) grew well on an array of complex and simple structures, while Bacteroides ovatus 3-1-23 grew well only on the simple structures. In a competition experiment, B. cellulosilyticus growth was favored over B. ovatus on the complex AX-based structure. On the other hand, on the simple structure, B. ovatus strongly outcompeted B. cellulosilyticus, which was eliminated from the competitive environment by Day 11. This adaptation to fine structure and resulting competition dynamics indicate that dietary fiber chemical structures, whether complex or simple, favor certain gut bacteria. Overall, this work supports a concept that fiber degraders diversify their competitive abilities to access substrates across the spectrum of heterogeneity of fine structural features of dietary fibers.