A novel DPH5-related diphthamide-deficiency syndrome causing embryonic lethality or profound neurodevelopmental disorder.

PURPOSE: Diphthamide is a post-translationally modified histidine essential for messenger RNA translation and ribosomal protein synthesis. We present evidence for DPH5 as a novel cause of embryonic lethality and profound neurodevelopmental delays (NDDs). METHODS: Molecular testing was performed using exome or genome sequencing. A targeted Dph5 knockin mouse (C57BL/6Ncrl-Dph5em1Mbp/Mmucd) was created for a DPH5 p.His260Arg homozygous variant identified in 1 family. Adenosine diphosphate-ribosylation assays in DPH5-knockout human and yeast cells and in silico modeling were performed for the identified DPH5 potential pathogenic variants. RESULTS: DPH5 variants p.His260Arg (homozygous), p.Asn110Ser and p.Arg207Ter (heterozygous), and p.Asn174LysfsTer10 (homozygous) were identified in 3 unrelated families with distinct overlapping craniofacial features, profound NDDs, multisystem abnormalities, and miscarriages. Dph5 p.His260Arg homozygous knockin was embryonically lethal with only 1 subviable mouse exhibiting impaired growth, craniofacial dysmorphology, and multisystem dysfunction recapitulating the human phenotype. Adenosine diphosphate-ribosylation assays showed absent to decreased function in DPH5-knockout human and yeast cells. In silico modeling of the variants showed altered DPH5 structure and disruption of its interaction with eEF2. CONCLUSION: We provide strong clinical, biochemical, and functional evidence for DPH5 as a novel cause of embryonic lethality or profound NDDs with multisystem involvement and expand diphthamide-deficiency syndromes and ribosomopathies.

Cumulus: A federated EHR-based learning system powered by FHIR and AI.

Objective: To address challenges in large-scale electronic health record (EHR) data exchange, we sought to develop, deploy, and test an open source, cloud-hosted app 'listener' that accesses standardized data across the SMART/HL7 Bulk FHIR Access application programming interface (API). Methods: We advance a model for scalable, federated, data sharing and learning. Cumulus software is designed to address key technology and policy desiderata including local utility, control, and administrative simplicity as well as privacy preservation during robust data sharing, and AI for processing unstructured text. Results: Cumulus relies on containerized, cloud-hosted software, installed within a healthcare organization's security envelope. Cumulus accesses EHR data via the Bulk FHIR interface and streamlines automated processing and sharing. The modular design enables use of the latest AI and natural language processing tools and supports provider autonomy and administrative simplicity. In an initial test, Cumulus was deployed across five healthcare systems each partnered with public health. Cumulus output is patient counts which were aggregated into a table stratifying variables of interest to enable population health studies. All code is available open source. A policy stipulating that only aggregate data leave the institution greatly facilitated data sharing agreements. Discussion and Conclusion: Cumulus addresses barriers to data sharing based on (1) federally required support for standard APIs (2), increasing use of cloud computing, and (3) advances in AI. There is potential for scalability to support learning across myriad network configurations and use cases.

Cumulus: a federated electronic health record-based learning system powered by Fast Healthcare Interoperability Resources and artificial intelligence.

OBJECTIVE: To address challenges in large-scale electronic health record (EHR) data exchange, we sought to develop, deploy, and test an open source, cloud-hosted app "listener" that accesses standardized data across the SMART/HL7 Bulk FHIR Access application programming interface (API). METHODS: We advance a model for scalable, federated, data sharing and learning. Cumulus software is designed to address key technology and policy desiderata including local utility, control, and administrative simplicity as well as privacy preservation during robust data sharing, and artificial intelligence (AI) for processing unstructured text. RESULTS: Cumulus relies on containerized, cloud-hosted software, installed within a healthcare organization's security envelope. Cumulus accesses EHR data via the Bulk FHIR interface and streamlines automated processing and sharing. The modular design enables use of the latest AI and natural language processing tools and supports provider autonomy and administrative simplicity. In an initial test, Cumulus was deployed across 5 healthcare systems each partnered with public health. Cumulus output is patient counts which were aggregated into a table stratifying variables of interest to enable population health studies. All code is available open source. A policy stipulating that only aggregate data leave the institution greatly facilitated data sharing agreements. DISCUSSION AND CONCLUSION: Cumulus addresses barriers to data sharing based on (1) federally required support for standard APIs, (2) increasing use of cloud computing, and (3) advances in AI. There is potential for scalability to support learning across myriad network configurations and use cases.

Predicting venous thromboembolism (VTE) risk in cancer patients using machine learning.

Background: The association between cancer and venous thromboembolism (VTE) is well-established with cancer patients accounting for approximately 20% of all VTE incidents. In this paper, we have performed a comparison of machine learning (ML) methods to traditional clinical scoring models for predicting the occurrence of VTE in a cancer patient population, identified important features (clinical biomarkers) for ML model predictions, and examined how different approaches to reducing the number of features used in the model impact model performance. Methods: We have developed an ML pipeline including three separate feature selection processes and applied it to routine patient care data from the electronic health records of 1910 cancer patients at the University of California Davis Medical Center. Results: Our ML-based prediction model achieved an area under the receiver operating characteristic curve of 0.778 ± 0.006 (mean ± SD) when trained on a set of 15 features. This result is comparable with the model performance when trained on all features in our feature pool [0.779 ± 0.006 (mean ± SD) with 29 features]. Our result surpasses the most validated clinical scoring system for VTE risk assessment in cancer patients by 16.1%. We additionally found cancer stage information to be a useful predictor after all performed feature selection processes despite not being used in existing score-based approaches. Conclusion: From these findings, we observe that ML can offer new insights and a significant improvement over the most validated clinical VTE risk scoring systems in cancer patients. The results of this study also allowed us to draw insight into our feature pool and identify the features that could have the most utility in the context of developing an efficient ML classifier. While a model trained on our entire feature pool of 29 features significantly outperformed the traditionally used clinical scoring system, we were able to achieve an equivalent performance using a subset of only 15 features through strategic feature selection methods. These results are encouraging for potential applications of ML to predicting cancer-associated VTE in clinical settings such as in bedside decision support systems where feature availability may be limited.

Ophthalmic manifestations in Costello syndrome caused by Ras pathway dysregulation during development.

BACKGROUND: Costello syndrome (CS) is a multisystem developmental disorder caused by germline pathogenic variants in HRAS resulting in dysregulation of the Ras pathway. A systematic characterization of ophthalmic manifestations provides a unique opportunity to understand the role of Ras signal transduction in ocular development and guide optimal ophthalmic care in CS individuals. METHODS: Visual function, ocular features and genotype/phenotype correlations were evaluated in CS individuals harboring HRAS pathogenic variants, by cross-sectional and retrospective studies, and were recruited through the Costello Syndrome Family Network (CSFN) between 2007 and 2020. RESULTS: Fifty-six molecularly diagnosed CS individuals including 34 females and 22 males, ages ranging from 0.5 to 37 years were enrolled. The most common ophthalmic manifestations in the cross-sectional study were lack of stereopsis (96%), refractive errors (83%), strabismus (72%), nystagmus (69%), optic nerve hypoplasia or pallor (55%) and ptosis (13.7%) with higher prevalence than in the retrospective data (refractive errors (41%), strabismus (44%), nystagmus (26%), optic nerve hypoplasia or pallor (7%) and ptosis (11%)). Visual acuities were found to ranged from 20/25 to 20/800 and contrast sensitivity from 1.6% to 44%. HRAS pathogenic variants included p.G12S (84%), p.G13C (7%), p.G12A (5.4%), p.G12C (1.8%) and p.A146V (1.8%). CONCLUSION: Majority of individuals with CS have refractive errors, strabismus, nystagmus, absent stereopsis, and optic nerve abnormalities suggesting that HRAS and the Ras pathway play a vital role in visual system development. Ptosis, refractive errors and strabismus are amenable to treatment and early ophthalmic evaluation is crucial to prevent long-term vision impairment and improve overall quality of life in CS.

Mucosal Healing in Clinical Practice: A Single-Center Pediatric IBD Experience.

BACKGROUND: Mucosal healing (MH) is associated with improved clinical outcomes in patients with Crohn's disease (CD) and ulcerative colitis (UC). MH as a target for treatment has been suggested, although there is little pediatric data. The goal of this study was to evaluate MH in clinical practice in pediatric patients with inflammatory bowel disease in clinical remission. METHODS: A retrospective review of electronic health record data was performed on all patients with CD or UC who underwent at least 2 colonoscopies from 2010 through 2016. Only patients in clinical remission undergoing a scope for MH were included in our study. The incidence of MH and histologic healing (HH) was analyzed, along with cumulative rates of MH in each group. MH was defined by both physician assessment of MH and an endoscopic score of zero for CD and UC. RESULTS: A total of 76 patients with CD and 28 patients with UC underwent at least one MH scope while in clinical remission. Of the 76 patients with CD, 51 patients (67%) demonstrated MH by physician assessment, 34 patients (45%) demonstrated MH by a simple endoscopic score for CD of zero, and 35 patients (46%) demonstrated HH. Of the 28 patients with UC, 20 patients (71%) demonstrated MH by physician assessment, 10 patients (36%) demonstrated MH by a Mayo score of zero, and 10 patients (36%) demonstrated HH. Nineteen patients underwent a second MH scope and 11 (58%) demonstrated MH by physician assessment, 7 patients (37%) demonstrated MH by simple endoscopic score for CD or Mayo scores of zero, and 5 patients (26%) demonstrated HH. Of those patients with active disease, 21 of 25 patients with CD underwent escalation of therapy, whereas 8 of 8 patients with UC underwent escalation of therapy. Cumulative rates of MH when defined by physician assessment were 79% (60 of 76 patients) in CD and 79% (22 of 28 patients) in UC. CONCLUSIONS: MH is feasible in pediatric CD and UC, and rates of cumulative MH in pediatric patients are similar to previously published adult data. In children with inflammatory bowel disease in clinical remission, approximately one-third demonstrate active disease at endoscopy.

Risk factors for quantity not sufficient sweat collection in infants 3 months or younger.

OBJECTIVES: The purpose is to identify demographic characteristics associated with a quantity not sufficient (QNS) sweat collection in infants 3 months or younger. METHODS: History of premature birth, infant race and sex, gestational age at delivery, and weight of the infant were compared with QNS collection. RESULTS: Of 221 sweat collections from 197 infants, 25 were QNS. Infant weight less than 3 kg and history of prematurity were associated with QNS collection (P < .001). Thirteen (30.2%) of 43 infants weighing less than 3 kg had QNS collections compared with 12 (7.9%) of 151 infants 3 kg or more. Twelve (46.2%) premature infants had QNS collections compared with 13 (7.6%) term infants. Lower birth gestational age and corrected gestational age were associated with QNS collections. Six (86%) of seven infants who weighed less than 3 kg, had a history of prematurity, and were more than 54 days old at testing had a QNS result. Sex and race did not correlate with QNS collections. CONCLUSIONS: Weight less than 3 kg and history of prematurity are associated with an increased chance of QNS sweat collections.

Encouraging adolescents toward mental health careers via website biographies.

This project designed, developed, implemented and is currently evaluating the effectiveness of an interactive, multi-media website designed to encourage adolescents to consider careers in mental health. This Web-based learning environment features biographies of mental health scientists. Evaluation is conducted in a systematic, structured way using cognitive achievement, usability (ease of use), and affective scales (e.g., fun to use) as outcome measures