The University of Michigan created the Practice-Oriented Research Training (PORT) program and implemented it between 2008 and 2018. The PORT program provided research training and funding opportunities for allied healthcare professionals. The program consisted of weekly didactics and group discussion related to topics relevant to developing specific research ideas into projects and funding for a mentored research project for those who submitted a competitive grant application. The goal of this evaluation was to assess the long-term impact of the PORT program on the research careers of the participants. Ninety-two participants (74 staff and 18 faculty) participated in both phases of the program. A mixed-methods approach to evaluation was used; 25 participants who received funding for their research completed surveys, and semi-structured interviews were conducted with eight program participants. In addition, data were collected on participants’ publication history. Fifteen out of the 74 staff participants published 31 first-authored papers after participating in PORT. Twelve out of 15 staff participants who published first-authored papers did so for the first time after participating in the PORT program. Results of quantitative and qualitative analyses suggest that the PORT program had positive impacts on both participants and the research community.

OBJECTIVES/SPECIFIC AIMS: The aim of this study is to determine whether quantitative measures of knee structures including effusion, bone marrow lesions, cartilage, and meniscal damage can improve upon an existing model of demographic and clinical characteristics to classify accelerated knee osteoarthritis (AKOA). METHODS/STUDY POPULATION: We conducted a case-control study using data from baseline and four annual follow-up visits from the osteoarthritis initiative. Participants had no radiographic knee osteoarthritis (KOA) at baseline. AKOA is defined as progressing from no KOA to advance-stage KOA in at least 1 knee within 48 months. AKOA knees were matched 1:1 based on sex to (1) participants who did not develop KOA within 48 months and (2) participants who developed KOA but not AKOA. Analyses were person based. Classification and regression tree analysis was used to determine the important variables and percent of variance explained. RESULTS/ANTICIPATED RESULTS: A previous classification and regression tree analysis found that age, BMI, serum glucose, and femorotibial angle explained 31% of the variability between those who did and did not develop AKOA. Including structural measurements as candidate variables yielded a model that included effusion, BMI, serum glucose, cruciate ligament degeneration and coronal slope and explained 39% of the variability. DISCUSSION/SIGNIFICANCE OF IMPACT: Knee structural measurements improve classification of participants who developed AKOA Versus those who did not. Further research is needed to better classify patients at risk for AKOA.

OBJECTIVES/SPECIFIC AIMS: The goal of this study was to investigate whether RF-RDN attenuates renal fibrosis and inflammation in SHR with established hypertension. METHODS/STUDY POPULATION: Twenty-two-week-old SHR received bilateral RF-RDN or Sham-RDN (Biosense Webster Stockert 70 generator and RF-probe). Four weeks later, SHR were sacrificed and paraffin sections of kidneys were stained for fibrosis by Masson’s trichrome staining. Kidney tissue were homogenized for measurement of cytokines levels by ELISA. RESULTS/ANTICIPATED RESULTS: The results showed that Sham-RDN treated SHR had extensive fibrosis as demonstrated by moderate thickening of Bowman’s capsule, collagen deposition in glomerulus, extensive tubulointerstitial fibrosis, and segmental glomerulosclerosis. In contrast, RF-RDN significantly reduced each of these pathological components of fibrosis in kidney cortex and medulla as compared with Sham-RDN treated kidneys. In other studies, RF-RDN decreased B cells, CD4+ T cells, and CD8+ T cells in the kidney of SHR as measured by flow cytometry. Meanwhile, kidney tissue levels of IL-17, INF-γ, MIP-3a, TNF-α, and TGF-β were decreased as compared with respective levels in Sham-RDN. DISCUSSION/SIGNIFICANCE OF IMPACT: Together, these findings demonstrate that removal of the influence of heightened renal sympathetic activity by RF-RDN decreases kidney inflammatory markers and attenuates renal fibrosis in hypertensive SHR.

Objectives/Goals: In patients with paraesophageal hernia (PEH), evaluate associations between CT radiomic features and clinical data with (1) undergoing elective repair and (2) volvulus/gastric outlet obstruction and develop an ML pipeline for automated feature extraction and prognosis estimation. Methods/Study Population: Retrospective case–control of adults (>18) at Penn Medicine (2017–2025) with hiatal hernia ICD-9/10 codes and CT chest or CT abdomen/pelvis from clinical care. Radiology reports, H&P, and operative notes will be used to identify those with PEH. We will use traditional statistics and classical ML methods (linear/logistic regression, decision trees, random forest) to test associations between radiomic features (hiatal defect diameter (HDD), hernia sac volume (HSV), herniated gastric volume (HGV)) and outcomes. We will explore CNN-based models – nnU-Net (e.g., TotalSegmentator) and ResNet-based models (e.g., Merlin) – and foundation models (MedSAM2, VISTA-3D) to build the pipeline. Results/Anticipated Results: We hypothesize that CT-derived radiomic features will be associated with (1) undergoing elective repair and (2) developing acute volvulus/GOO. Greater HDD, HSV, and HGV are expected to be associated with increased odds of both outcomes. The ML pipeline is expected to provide organ segmentation and reproducible automated feature extraction using state-of-the-art architectures, and to capture synergistic effects between radiographic and clinical variables to inform patient-level prognosis. Discussion/Significance of Impact: In asymptomatic/minimally symptomatic patients, PEH care often defaults to watchful waiting; candidacy and timing for elective repair are uncertain. Our ML pipeline integrates clinical and CT data to standardize preop radiomics, estimate prognosis, and inform if/when to operate.

Objectives/Goals: Using secure systems for sharing documents with external collaborators is essential for all researchers. These documents may include protected health information (PHI) or sensitive materials like protocols, study reports, DSMB reports, publications, presentations, abstracts, and statistical analysis plans (SAPs). Methods/Study Population: We surveyed the ACTS Biostatistics, Epidemiology, and Research Design Special Interest Group (BERD-SIG) to gather information about the systems they are currently using or have used in the past for document sharing with external collaborators. The survey focused on the security of these systems, particularly in relation to sharing documents containing PHI. In addition, the survey included questions about various system features of interest. These features included version control, simultaneous editing by multiple users, and access rights management, such as the ability to assign different permissions (e.g., read-only, write, and download) to different individuals. We also invited participants to provide feedback on any additional positive or negative aspects of the systems they use. Results/Anticipated Results: We received 28 completed survey responses. Respondents had an option for choosing more than one system. The top current systems reported were Microsoft Teams (OneDrive, SharePoint) (n = 16), Box (n = 11), Google Docs/Drive (n = 10), and Dropbox (n = 6). Among other systems listed individually were Filelocker, REDCap, Slack, Website, Significant Media Shuttle, and Zulip. Notably, 15 responses indicated the respondents were unsure if their system is secure for sharing documents containing PHI. Respondents also offered feedback on both the positive and negative aspects of these systems. For example, a key advantage of Box was its password-controlled access. However, its incompatibility with office tools and the challenges for external collaborators attempting to access the system were noted as drawbacks. Discussion/Significance of Impact: Utilizing secure institutional document-sharing systems and understanding their features significantly affects the effectiveness and security of collaborations among researchers, particularly with external partners. This knowledge is especially crucial when sharing documents containing sensitive patient and study data.

ABSTRACT IMPACT: This study provides public health and K-12 school districts with a pragmatic, flexible, adaptable model showing COVID-19 transmission dynamics, using local data and program elements that are modifiable and with an online model for easy use, to enable safe and equitable re-opening and maintenance of in-person learning. OBJECTIVES/GOALS: School closures resulting from the COVID-19 pandemic disrupt student education and health and exacerbate inequities. Public health agencies and school districts currently lack pragmatic models to assess the effects of potential strategies for resuming and maintaining in-person learning on outcomes such as transmission and attendance. METHODS/STUDY POPULATION: This study explored how various combinations of transmission-mitigating interventions affect health and learning outcomes in a range of underlying epidemiological conditions. The CTSA science team developed a conceptual framework and an agent-based simulation model with parameters including prevalence, transmission, testing, preventive and responsive actions, infection control, population behavior and awareness, and the potential impact of vaccine adoption and exemption policies. The team partnered with a large school district to ensure relevance of the program components to decision-making. RESULTS/ANTICIPATED RESULTS: The model shows that no single program element or condition ensures safety. Combining interventions can result in synergy in the mitigation efforts. Even without testing, an efficient health screening process with forthcoming risk reporting, combined with on-campus infection control, can reduce on-campus transmission. The resulting model is accessible online to enable exploration of likely scenarios. It is adaptable as COVID-19 science evolves, including for testing and vaccines. DISCUSSION/SIGNIFICANCE OF FINDINGS: This research provides public health agencies and school districts with a model that couples local conditions with programmatic elements to help inform the local COVID-19 response, recognizing that decisions about the school community are often complex politically, technically, and operationally when it comes to addressing a health crisis.

OBJECTIVES/SPECIFIC AIMS: Increased retinal protein O-GlcNAcylation occurs in response to hyperglycemia and contributes to diabetic retinopathy. Renin-angiotensin system (RAS) blockers reduce the incidence of diabetic retinopathy. Beneficial effects of RAS blockers are often attributed to production of angiotensin-(1-7) (Ang1-7). The objective here is to determine the impact of Ang1-7 on retinal protein O-GlcNAcylation. METHODS/STUDY POPULATION: C57/BL6 mice were fed a high-fat diet for 8 weeks and then treated for 3 weeks with either a vehicle control, the RAS blocker captopril, or captopril and the Ang1-7 receptor antagonist A779. R28 cells were used to assess levels of O-GlcNAcylated proteins in response to Ang1-7, and the role of cAMP was investigated with addition of forskolin, 6-Bnz-cAMP-AM, and 8-pCPT-2-O-Me-cAMP-AM to cell culture medium. RESULTS/ANTICIPATED RESULTS: Captopril attenuated retinal protein O-GlcNAcylation in mice fed a high-fat diet. This effect was reversed by A779. Ang1-7 attenuated protein O-GlcNAcylation and increased cAMP levels. Forskolin and the EPAC selective cAMP analog 8-pCPT-2-O-Me-cAMP-AM, but not the PKA selective cAMP analog 6-Bnz-cAMP-AM, attenuated O-GlcNAcylation. Inhibiting EPAC blocked the effect of forskolin, whereas inhibiting PKA did not. DISCUSSION/SIGNIFICANCE OF IMPACT: This study demonstrates a novel role for Ang1-7 in the retina and identifies a potential EPAC-dependent mechanism that regulates protein O-GlcNAcylation. Thus, future therapeutics targeted at an Ang1-7/EPAC axis in retina may be used to address DR.

OBJECTIVES/GOALS: We were interested in health equity research for each CTSA-affiliated institution, specifically focusing on cross department and cross-campus co-authorship. We conducted a bibliometric analysis of our CTSA-funded papers relating to diversity and inclusion to identify cross department and cross-campus collaborations. METHODS/STUDY POPULATION: We worked with our CTSAs Racial Justice, Diversity, Equity and Inclusion Task Force to conduct an environmental scan of diversity and inclusion research across our CTSA partner institutions. Using the Scopus database, searches were constructed to identify and retrieve the variety of affiliations for each of the CTSA authors, a health equity/health disparities search hedge, and all of our CTSA grant numbers. We limited the dates from the beginning of our CTSA in 2008-November 2021. We used PubMed to retrieve all MeSH terms for the articles. We used Excel to analyze the data, Python and NCBIs Entrez Programming Utilities to analyze MeSH terms, and VOSviewer to produce the visualizations. RESULTS/ANTICIPATED RESULTS: The results of this search yielded 94 articles overall. We broke these up into subsets (not mutually exclusive) to represent five of the researcher groups across our CTSA. We analyzed the overall dataset for citation count, normalized citation count, CTSA average authors, gender trends, and co-term analysis. We also developed cross department co-authorship maps and cross-institutional/group co-authorship maps. DISCUSSION/SIGNIFICANCE: This poster will demonstrate both the current areas where cross-departmental and cross-institutional collaboration exists among our CTSA authors, as well as identify potential existing areas for collaboration to occur. These findings may determine areas our CTSA can support to improve institutional performance in addressing health equity.

OBJECTIVES/GOALS: Observational studies suggest unequal effects of COVID-19 on the population of the U.S. distinguished by race and ethnicity. Our primary research question: what are the demographic differences among patients identified with concurrent ischemic stroke and COVID-19 positivity? METHODS/STUDY POPULATION: The National Covid Cohort Collaboration (N3C) data was used to identify patients with concurrent COVID-19 and stroke, operationally defined as those with a COVID diagnosis and inpatient admission for ischemic stroke 1 week before or 6 weeks after their COVID diagnosis. The data was further age restricted (18-65 years) and a categorical variable was created representing payer plans (Medicaid, Medicare, Other insurance). Data on patients race/ethnicity, comorbidities, treatments administered (Remdesivir and ECMO) and insurance information was analyzed using various exploratory data methods and visualizations. Logistic regression was implemented to model the relationship between variables (dependent/independent) in the cohorts. Model complexity was analyzed using the F test of significance. RESULTS/ANTICIPATED RESULTS: Taken as a whole, the data contained over 7 billion rows and around 6.4 million persons (~ 2.15 million of whom were COVID+). The main cohort of individuals with concurrent COVID positivity and ischemic stroke made up around 0.29% of the original COVID+ group, and the payer plan sub-cohort consists of around 29.26% of our main cohort. Black/African American (AA) and the Hispanic/Latino any Race have younger distributions (median ~ 65 years), while the White Non-Hispanic group has the oldest distribution (median ~ 70 years). Black/AA had the highest average number of comorbidities per patient (4.49), compared to white non-Hispanic (3.39) and Asian non-Hispanic (2.59). In our analysis, Medicaid patients had lower odds of obtaining ECMO (p < .01), there was no significant difference in Remdesivir treatment. DISCUSSION/SIGNIFICANCE: We found the N3C data to be useful in studying a distinct group of patients, and exploring COVID-19 and ischemic stroke treatment across patients’ race/ethnicity identities and insurance status. Our exploratory analysis provides a foundation for further insight into demographic trends and discrepancies in apportionment of treatment.

OBJECTIVES/SPECIFIC AIMS: Selective loss of long-projecting neural circuitry is a common feature of many neurodegenerative diseases, such as the vulnerable nigrostriatal pathway in Parkinson’s disease (PD). Current in vitro approaches for studying disease development generally do not mimic complex anatomical features of the afflicted substrates such as long axonal pathways between stereotypical neural populations. Such exquisite features are not only crucial for neural systems function but may also contribute to the preferential vulnerability and pathophysiological progression of these structures in neurodegenerative disease. We have previously developed micro-tissue engineered neural networks to recapitulate the anatomy of long-projecting cortical axonal tracts encased in a tubular hydrogel.1 Recently, we have extended this work to include the first tissue-engineered nigrostriatal pathway that was anatomically-inspired to replicate the structure and function of the native pathway.2 Notably, this tissue-engineered brain pathway possesses three-dimensional (3D) structure, multicellular composition, and architecture of long axonal tracts between distinct neuronal populations. Therefore, in the current study we apply this system as a biofidelic test-bed for evaluating axonal pathway development, maturation, and pathophysiology. METHODS/STUDY POPULATION: Dopaminergic neurons from the ventral mesencephalon and medium spiny neurons (MSNs) from the striatum were separately isolated from rat embryos. Tissue-engineered nigrostriatal pathways were formed by initially seeding dopaminergic neuron aggregates at one end of hollow hydrogel micro-columns with a central extracellular matrix, collectively spanning up to several centimeters in length. Several days later, tissue-engineered MSN aggregate was seeded on the other end and was allowed to integrate. Immunocytochemistry (ICC) and confocal microscopy were used to assess health, cytoarchitecture, synaptic integration, and mitochondrial dynamics with stains that label cell nuclei (Hoechst) and mitochondria (MitoTracker Red) and antibodies that recognize axons (anti-β-tubulinIII), neurons/dendrites (anti-MAP2), dopaminergic neurons/axons (anti-tyrosine hydroxylase; TH), and MSNs (anti-DARPP-32). RESULTS/ANTICIPATED RESULTS: Seeding tubular micro-columns with dopaminergic neuronal aggregates resulted in unidirectional axonal extension, ultimately spanning >5mm by 14 days in vitro. For constructs also seeded with Tissue-engineered, ICC confirmed the presence of the appropriate neuronal sub-types in the two aggregate populations, specifically TH+ dopaminergic neurons and DARPP-32+ MSNs. Moreover, confocal microscopy revealed extensive axonal-dendritic integration and synapse formation involving the dopaminergic axons and MSN somata/dendrites. Collectively, these constructs mimicked the general cytoarchitecture of the in vivo nigrostriatal pathway: a discrete population of dopaminergic neurons with long-projecting 3D axonal tracts that were synaptically integrated with a population of MSNs. Mitochondria structure along axonal tracts was also observed using MitoTracker staining, revealing dynamic intra-axonal mitochondrial motility in this system. Ongoing studies are evaluating real-time mitochondrial dynamics and axonal physiology in this tissue-engineered nigrostriatal pathway in vitro, under both baseline conditions as well as following the addition of exogenous α-Synuclein fibrils to model synucleinopathy in PD. DISCUSSION/SIGNIFICANCE OF IMPACT: This tissue-engineered nigrostriatal pathway provides an anatomically-inspired platform with neuronal-axonal architecture that structurally and functionally emulates the nigrostriatal pathway in vivo. We are applying this paradigm as a powerful in vitro test-bed for understanding mitochondrial activity and inter-axonal energetics pathways under homeostatic as well as PD pathological conditions. Successful demonstration will serve as proof-of-concept that this technique can be used to study mitochondria pathology in personalized constructs built using cells derived from PD patients in order to evaluate pharmacological therapies targeted at improving mitochondrial resiliency and fitness so as to delay and/or prevent dopaminergic axonal/neuronal degeneration in tailored to specific PD patients.

OBJECTIVES/SPECIFIC AIMS: Urinary tract infections (UTIs) serve as one of the most common infections affecting women. With rising reports of antibiotic resistance (ABR), which can prolong illness and limit treatment options, the Infectious Disease Society of America recommends using local resistance patterns to shape empirical treatment selection. Although no studies have evaluated ABR in Ureaplasma spp. urinary isolates in college-aged women, regional studies in the Southeast United States have found levels of tetracycline resistance in over 30% of Ureaplasma spp. clinical isolates. Thus, this study aims to determine the antibiogram for 73 Ureaplasma spp. and 10 Mycoplasma hominis isolates collected from women with first-time UTI against a panel of 9 antibiotics, and assess resistant isolates for genetic mechanisms associated with resistance. METHODS/STUDY POPULATION: This study used archival samples and data collected from college-aged women with first-time UTI recruited to participate in a prospective cohort study conducted at a student healthcare facility from 2001 to 2006 in Florida. Ureaplasma spp. and M. hominis isolates cultured from urine samples collected at the initial clinical presentation and for any recurrent UTI were evaluated for susceptibility to a panel of 9 antibiotics (8 for M. hominis) using validated microbroth and agar dilution methods, respectively. Ureaplasma spp. isolates were tested against azithromycin, chloramphenicol, ciprofloxacin, clindamycin, erythromycin, doxycycline, gentamicin, levofloxacin, and tetracycline. M. hominis isolates underwent the same testing, with the addition of linezolid and exclusion of azithromycin and erythromycin, as M. hominis is intrinsically resistance to 14 and 15-membered macrolides and azilides. PCR and Sanger sequencing were employed to identify molecular mechanisms associated with resistance. RESULTS/ANTICIPATED RESULTS: Of the 73 Ureaplasma spp. isolates, 1 isolate was resistant to levofloxacin (MIC: 4 µg/mL) and 1 to tetracycline (MIC: 8 µg/mL). All M. hominis isolates were sensitive. For the Ureaplasma spp. isolates, MIC90s were highest against gentamicin (32 µg/mL) and lowest against doxycycline (0.25 µg/mL). PCR amplification identified tetM present in the tetracycline resistant isolate, an established gene associated with tetracycline resistance in Ureaplasma spp. A S83W mutation within the quinolone-resistance-determining region (QRDR) of parC was detected in the levofloxacin resistant isolate. DISCUSSION/SIGNIFICANCE OF IMPACT: Overall, antibiotic resistance in this population of college-aged women with first-time UTI was low. A previous study detected a novel S83W substitution in a perinatal Ureaplasma spp. isolate from Japan, and provided in silico evidence that a S83W change would prevent levofloxacin from binding to its target. However, that study was unable to cultivate the isolate. Our study has provided the corresponding phenotypic evidence that a S83W substitution results in quinolone resistance in Ureaplasma spp.

OBJECTIVES/GOALS: Computational pathology is an emerging discipline that resides at the intersection of engineering, computer science, and pathology. There is a growing need to develop innovative pedagogical approaches to train future computational pathologists who have diverse educational backgrounds. METHODS/STUDY POPULATION: Our work proposes an iterative approach toward teaching master’s and Ph.D. students from various backgrounds, such as electrical engineering, biomedical engineering, and cell biology the basics of cell-type identification. This approach is grounded in the active learning framework to allow for observation, reflection, and independent application. The learners are trained by a team of an electrical engineer and pathologist and provided with eight images containing a glomerulus. They must then classify nuclei in each of the glomeruli as either a podocyte (blue), endothelial cell (green), or mesangial cell (red). RESULTS/ANTICIPATED RESULTS: A simple web application was built to calculate agreement, measured using Cohen’s kappa, between annotators for both individual glomeruli and across all eight images. Automating the process of providing feedback from an expert renal pathologist to the learner allows for learners to quickly determine where they can improve. After initial training, agreement scores for cells scored by both the learner and the expert were high (0.75), however, when including cells not scored by both the agreement was relatively low (0.45). This indicates that learners needed more instruction on identifying unique cells within each image. This low-stakes approach encourages exploratory and generative learning. DISCUSSION/SIGNIFICANCE: Computation medical sciences require interdisciplinary training methods. We report on a robust approach for team-based mentoring and skill development. Future implementations will include undergraduate learners and provide opportunities for graduate students to engage in near-peer mentoring.

OBJECTIVES/GOALS: Greater cognitive function (CF) is associated with adherence to prescription medications, better program adherence and weight loss (WL) following bariatric surgery. The purpose of this study was to evaluate the association between baseline CF, intervention dropout, adherence and 3-month WL. METHODS/STUDY POPULATION: 107 (Mage = 40.9 yrs.), overweight/obese (BMI = 35.6 kg/m2) men (N = 17) and women (N = 90) completed a 3-mo. WL intervention. Participants were asked to attend weekly behavioral sessions, comply with a reduced calorie diet and complete 100 min of moderate intensity physical activity (PA)/wk. CF tasks including Flanker (attention), Stroop (Executive control) and working memory, body weight and cardiovascular fitness (covariate) were assessed at baseline and 3-mos. Session attendance, adherence to PA and diet prescriptions and number of off-diet episodes were recorded weekly. RESULTS/ANTICIPATED RESULTS: Results indicated that attention was positively correlated with session attendance (p = 0.016), adherence to the diet (p < 0.01) and PA (p = 0.023). Executive control was positively correlated with WL (p = 0.042). Working memory (two tasks) was also positively correlated with WL (p = .017 and p = .025). Analysis of variance (ANOVA) indicated that baseline attention (p = .012) was positively related to WL and negatively associated with drop out (p < .05). Hierarchical linear regression showed executive control (p = .036, R2 = .054) and working memory (p = .013, R2 = .073 and p = .017, R2 = .068) were associated with WL when controlling for fitness. DISCUSSION/SIGNIFICANCE OF IMPACT: These results suggest that stronger baseline attention is associated with completion of a 3-mo. WL intervention. Executive control and working memory are associated with amount of WL achieved. Additional, larger and longer trials to assess the role of baseline CF on WL and evaluating the impact of interventions designed to improve CF on WL are indicated.

ABSTRACT IMPACT: We developed a novel semi-automated computational platform to quantify spatial characteristics of preserved peripheral visual field loci in retinitis pigmentosa. This work may inform future research on therapies to prevent visual field loss in retinitis pigmentosa and related diseases. OBJECTIVES/GOALS: Retinitis pigmentosa (RP) causes progressive and severe peripheral visual field (pVF) loss in some but not all patients. The characteristics of pVF in RP are incompletely understood. We developed a novel semi-automated computational platform to quantify the spatial characteristics of pVF. METHODS/STUDY POPULATION: We analyzed preserved pVF size and location in both eyes of RP patients using the Goldmann V4e isopter. We developed a custom algorithm in MATLAB to align and average the binary V4e isopter segmentations, and generated a two-dimensional probability map of the spatial distribution of preserved pVF loci along the radial and circumferential dimensions. To adjust for disease duration, cases were categorized by the time from self-reported symptom onset. Probability maps of pVF preservation were generated for categories of disease duration using unsupervised K-means clustering. Analyzing cases with longitudinal data, we identified loci of stable pVF over time. RESULTS/ANTICIPATED RESULTS: A total of 152 patients were included (N=304 eyes). The mean age was 46.7 years and 49.3% were male. Disease duration was categorized as <20 years (N=72, 47.4%), 20-40 years (N=60, 39.5%), or >40 years (N=20, 13.2%). Longitudinal data (3.2 -5.7 years of follow-up) were available in 65 patients (42.8%). Probability plots of preserved pVF loci in the cross-sectional dataset showed that the median percentage of preserved pVF loci were located between 50 ºand 80 ºeccentricity and between the 30 ºto 50 ºmeridians, with highly concordant inter-ocular symmetry. Probability plots in the longitudinal dataset showed that inferotemporal pVF loci were most likely to be preserved over time. DISCUSSION/SIGNIFICANCE OF FINDINGS: Semi-automated quantification of pVF loci is a useful platform to analyze spatial characteristics of the visual field in RP. Certain portions of the pVF may be relatively resistant to functional decline. Understanding the molecular basis of pVF resilience will inform further research on RP therapy.

OBJECTIVES/GOALS: An estimated 87% of torture survivors, or 27 million people globally, suffer from chronic pain such as brachial plexopathy from upper extremity suspension or lumbosacral plexus injury from leg hyperextension. However, a vast majority of pain is undetected by evaluators due to a lack of diagnostic tools and confounding psychiatric illness. This diagnostic gap results in exclusive psychological treatment rather than multimodal therapies, substantially limiting rehabilitation, placing vulnerable individuals at higher risk of drug abuse, and increasing healthcare expenditures. We hypothesized that the novel application in torture survivors of a validated pain screen can supplement the UNIP and improve its sensitivity for pain from approximately 15% to 90%, as compared to the reference standard. METHODS/STUDY POPULATION: In this prospective, blind comparison to gold standard study, 25 patients who survived torture by World Medical Association definition first received the standard evaluation protocol for torture survivors (United Nations Istanbul Protocol, UNIP) by a trained evaluator, and subsequently received a validated pain screen (Brief Pain Inventory Short Form, BPISF) followed by a non-invasive examination by a pain specialist physician (reference standard). The primary outcome was the diagnostic and treatment capability of the standard protocol (index test) versus the validated pain screen (BPISF), as compared to the reference standard. RESULTS/ANTICIPATED RESULTS: Providers using only the UNIP detected and treated pain in a maximum of 16% of patients as compared to 85% of patients who were diagnosed with pain by the reference standard. When employed, the validated pain screen had a sensitivity of 100% [95% CI: 72% - 100%] and a negative predictive value of 100%, as compared to a sensitivity of 24% [95% CI: 8% - 50%] and negative predictive value of 19% by the index test. The difference in the sensitivity of the UNIP as compared to the BPISF was significant, with p < 0.001. DISCUSSION/SIGNIFICANCE OF IMPACT : These data indicate that the current global standard assessment of torture survivors, the United Nations Istanbul Protocol, should be supplemented by the use of a validated pain screen to increase the accuracy of chronic pain diagnosis. This would change the standard medical assessment of 70.8 million people globally, a number that continues to rise by nearly 45,000 people each day. ClinicalTrials.gov protocol number NCT03018782.

OBJECTIVES/SPECIFIC AIMS: In June 2016, the FDA cautioned against the use of SGLT-2 inhibitors because of increased risk of acute kidney injury (AKI) after 101 cases of AKI were reported between March 2013 and October 2015. This study seeks to determine risk of AKI associated with SGLT-2 inhibitors in a large cohort of type 2 diabetic patients. METHODS/STUDY POPULATION: Retrospective cohort study including SGLT-2 inhibitor users and nonusers in the Mount Sinai Chronic Kidney Disease Registry between January 2013 and September 2016. SGLT-2 inhibitor users and nonusers were type 2 diabetics with new SGLT-2 inhibitor prescription after January 2013 and an outpatient visit between 2013 and 2015, respectively. Subjects were propensity matched by nearest neighbor method based on demographics, comorbidities, laboratory values, medications, estimated glomerular filtration rate (eGFR), and length of follow-up. The primary end point was AKI (defined by KDIGO laboratory algorithm) occurring during the follow-up period. RESULTS/ANTICIPATED RESULTS: In total, 372 SGLT-2 inhibitor users [mean age 63 years; 205 (55%) men] and 372 [mean age 63; 194 (52%) men] nonusers were included in the primary analysis. Proportions of AKI events defined by KDIGO criteria in users and nonusers were 4.0% and 10.0%, respectively. Adjusted odds ratio for AKI was 1.00 (95% CI, 0.28–2.62). Median peak serum creatinine measurements during AKI events for user and nonuser groups were 1.60 (IQR 1.36–1.78) and 1.88 (IQR 1.55–2.44) (p=0.02), respectively. Sensitivity analyses yielded similar results. DISCUSSION/SIGNIFICANCE OF IMPACT: These findings suggest that there is no evidence of increased odds of AKI in SGLT-2 inhibitor users compared with propensity-matched nonusers with type 2 diabetes.

OBJECTIVES/GOALS: Incorporating health equity, diversity, and inclusion (HEDI) development during community advisory board meetings is essential for ensuring that perspectives of all community members are considered, that health research is centered on the experiences of historically marginalized groups, and organizational strategies align with the community. METHODS/STUDY POPULATION: All IN for Health is a public engagement program that promotes health and research literacy, seeking to increase the public’s understanding of the role and value of research in improving health. It is guided by an active CAB providing advice on strategic directions, feedback to all efforts, contributing ideas, priorities, and most importantly, accountability. In 2023, ALL IN for Health started to incorporate HEDI professional development during staff team meetings and most recently, into our quarterly CAB meetings. Initially, the CAB was asked to provide feedback about talks and potential presentations related to health-research, sponsored, and shared by All IN for Health. During this exercise, board members were asked to provide HEDI topics of interest. Their responses informed a plan for team and board members. RESULTS/ANTICIPATED RESULTS: Some of the topics suggested by board members include: understanding health equity in relation to research studies and protocols, and topics specific to health research such as: health care access for rural areas and vulnerable populations, culture-based attitudes and beliefs and how they impact decisions related to health care, being aware of limitations certain communities have and what they may not have access to. The initiative was positively received and unanimously adopted. We hope to introduce the HEDI integration model for CABs at this conference. (Conceptual Model attached). DISCUSSION/SIGNIFICANCE: It is important to learn and grow alongside our community members. Such practice is bound to sustain partnerships that promote health equity, and exemplify meaningful community engagement, bidirectional learning, and a shared leadership model. By consistently incorporating and prioritizing HEDI, HAB can contribute to more equitable initiatives.

Objectives/Goals: To identify real-world type 2 diabetes (T2D) treatment pathways and evaluate whether AI models trained on electronic health record (EHR) medication histories can improve prediction of glycemic outcomes in routine clinical practice. Methods/Study Population: We conducted a retrospective cohort study using de-identified EHR data from the TriNetX Research Network (>3 M adults with T2D, 2019–2024). Medication, laboratory, and demographic data were standardized in PostgreSQL on the University of Utah CHPC. Patients with T1D, pregnancy, or incomplete follow-up were excluded. Longitudinal medication sequences were clustered using hierarchical agglomerative methods to identify treatment trajectories. Transformer-based models incorporating full prescription histories were trained to predict 6-month HbA1c, and performance (RMSE) was assessed overall and across demographic groups. Results/Anticipated Results: Among >3 M adults with T2D, we identified 40 distinct treatment clusters representing mono-, dual-, complex-, GLP-1–based, and variant pathways. GLP-1/GIP agents were typically used as 2nd–3rd-line therapy after metformin or SGLT2i and produced the largest BMI and HbA1c improvements. Baseline XGBoost models using EHR features predicted 6-month HbA1c with RMSE ≈ 0.72. The last HbA1c value was the most predictive feature, though additional labs and demographics modestly improved accuracy. Ongoing work is developing transformer models to capture longitudinal patterns and medication history effects. Discussion/Significance of Impact: National EHR networks can support development of trustworthy, fairness-aware AI to model diabetes pharmacotherapy. Insights reveal disparities in advanced therapy uptake and inform equitable decision-support design.

OBJECTIVES/SPECIFIC AIMS: The objective of this study is to measure the association of CYP2C19 (*1-*8,*17), ABCB1(C3435T; rs1045642), PON1 (p.Q192R; rs662), and B4GALT2 (c.909 C>T and c.366 G>C) gene polymorphisms in the Caribbean Hispanic population with major adverse cardiovascular events (MACE). METHODS/STUDY POPULATION: Patients of Caribbean Hispanic ethnicity from all geographic regions of the Island of Puerto Rico, male and female, aged >21 will be recruited. Cases will consist of patients receiving a daily clopidogrel dose of 75 mg following acute coronary syndrome (ACS) who experience a MACE within the first year of treatment. Control study patients must have received clopidogrel 75 mg daily for a minimum of 1 year without experiencing MACE. Genomic DNA samples will be genotyped to determine the frequency distribution of major CYP2C19, ABCB1, PON1, and B4GALT2 gene polymorphisms. Observed frequencies will be compared with other reported populations. An association study will be performed between genetic variables and MACE and a multivariable logistic regression model (additive) will be constructed. RESULTS/ANTICIPATED RESULTS: We anticipate finding a significant association between major genetic determinants of clopidogrel response and MACE where cases with MACE will carry higher frequency of CYP2C19, ABCB1, PON1, and B4GALT2. DISCUSSION/SIGNIFICANCE OF IMPACT: As the range of multiloci allelic combinations in admixed Caribbean Hispanics is higher than in other populations due to its unique 500-year history of genomic admixture, a wide spectrum of genetic variances is expected to be present in the study population. Determining the prevalence and effect of CYP2C19, ABCB1, PON1, and B4GALT2 polymorphisms holds the potential to personalize anti-platelet treatment for Caribbean Hispanic patients requiring treatment after ACS.

OBJECTIVES/GOALS: The design and phased roll-out of a CTSA competency-based longitudinal Clinical and Translational Research (CTR) curriculum pathway that can be integrated into the training of health sciences professional degree programs at three regional institutions. The outcome will be an increased number of health science professionals participating in CTR. METHODS/STUDY POPULATION: Based on an environmental scan, student surveys to assess interest, and feedback from regional stakeholders, a CTR pathway program was developed. The pathway curriculum will be grounded in 6 key focus areas derived from the Core Competencies in Clinical and Translational Research and prioritized based on regional need. The CTR pathway is currently being developed for the University of Washington School of Medicine, with additional implementations at the Washington State University School of Pharmacy and the Montana State University School of Nursing. Students will complete training modules and a mentored research project that is integrated into their specific course of study. RESULTS/ANTICIPATED RESULTS: In addition to the initial assessments and curriculum, an Advisory Committee will be established. Mentors, site leads, and research project partnerships across the region will be identified. Modifications will be made according to the local needs at both Montana State University and Washington State University. After the pilot launch and roll-out, the pathway curriculum will be adapted for other disciplines based on input from content experts and pathway evaluation data. Student retention in CTR fields will be tracked, with a goal to increase the number of CTR investigators and professionals across WWAMI in the next decade. CTR pathway processes and training resources will be shared with the CTSA consortium and other health sciences professional training sites. DISCUSSION/SIGNIFICANCE: Development of the CTR workforce is a priority to increase national capabilities in clinical and translational science. Building on a recognized need for targeted and longitudinal engagement, a CTR pathway is being established for health sciences students in the WWAMI region.