Progressive Metabolic Abnormalities Associated with the Development of Neonatal Bronchopulmonary Dysplasia

Objective: To assess the longitudinal metabolic patterns during the evolution of bronchopulmonary dysplasia (BPD) development. Methods: A case-control dataset of preterm infants (<32-week gestation) was obtained from a multicenter database, including 355 BPD cases and 395 controls. A total of 72 amino acid (AA) and acylcarnitine (AC) variables, along with infants’ calorie intake and growth outcomes, were measured on day of life 1, 7, 28, and 42. Logistic regression, clustering methods, and random forest statistical modeling were utilized to identify metabolic variables significantly associated with BPD development and to investigate their longitudinal patterns that are associated with BPD development. Results: A panel of 27 metabolic variables were observed to be longitudinally associated with BPD development. The involved metabolites increased from 1 predominant different AC by day 7 to 19 associated AA and AC compounds by day 28 and 16 metabolic features by day 42. Citrulline, alanine, glutamate, tyrosine, propionylcarnitine, free carnitine, acetylcarnitine, hydroxybutyrylcarnitine, and most median-chain ACs (C5:C10) were the most associated metabolites down-regulated in BPD babies over the early days of life, whereas phenylalanine, methionine, and hydroxypalmitoylcarnitine were observed to be up-regulated in BPD babies. Most calorie intake and growth outcomes revealed similar longitudinal patterns between BPD cases and controls over the first 6 weeks of life, after gestational adjustment. When combining with birth weight, the derived metabolic-based discriminative model observed some differences between those with and without BPD development, with c-statistics of 0.869 and 0.841 at day 7 and 28 of life on the test data. Conclusions: The metabolic panel we describe identified some metabolic differences in the blood associated with BPD pathogenesis. Further work is needed to determine whether these compounds could facilitate the monitoring and/or investigation of early-life metabolic status in the lung and other tissues for the prevention and management of BPD.

Identification of elders at higher risk for fall with statewide electronic health records and a machine learning algorithm

OBJECTIVE

Predicting the risk of falls in advance can benefit the quality of care and potentially reduce mortality and morbidity in the older population. The aim of this study was to construct and validate an electronic health record-based fall risk predictive tool to identify elders at a higher risk of falls.

METHODS

The one-year fall prediction model was developed using the machine-learning-based algorithm, XGBoost, and tested on an independent validation cohort. The data were collected from electronic health records (EHR) of Maine from 2016 to 2018, comprising 265,225 older patients (≥65 years of age).

RESULTS

This model attained a validated C-statistic of 0.807, where 50 % of the identified high-risk true positives were confirmed to fall during the first 94 days of next year. The model also captured in advance 58.01 % and 54.93 % of falls that happened within the first 30 and 30-60 days of next year. The identified high-risk patients of fall showed conditions of severe disease comorbidities, an enrichment of fall-increasing cardiovascular and mental medication prescriptions and increased historical clinical utilization, revealing the complexity of the underlying fall etiology. The XGBoost algorithm captured 157 impactful predictors into the final predictive model, where cognitive disorders, abnormalities of gait and balance, Parkinson's disease, fall history and osteoporosis were identified as the top-5 strongest predictors of the future fall event.

CONCLUSIONS

By using the EHR data, this risk assessment tool attained an improved discriminative ability and can be immediately deployed in the health system to provide automatic early warnings to older adults with increased fall risk and identify their personalized risk factors to facilitate customized fall interventions.

Improved detection of prostate cancer using a magneto-nanosensor assay for serum circulating autoantibodies

Purpose

To develop a magneto-nanosensor (MNS) based multiplex assay to measure protein and autoantibody biomarkers from human serum for prostate cancer (CaP) diagnosis.

Materials and methods

A 4-panel MNS autoantibody assay and a MNS protein assay were developed and optimized in our labs. Using these assays, serum concentration of six biomarkers including prostate-specific antigen (PSA) protein, free/total PSA ratio, as well as four autoantibodies against Parkinson disease 7 (PARK7), TAR DNA-binding protein 43 (TARDBP), Talin 1 (TLN1), and Caldesmon 1 (CALD1) and were analyzed. Human serum samples from 99 patients (50 with non-cancer and 49 with clinically localized CaP) were evaluated.

Results

The MNS assay showed excellent performance characteristics and no cross-reactivity. All autoantibody assays showed a statistically significant difference between CaP and non-cancer samples except for PARK7. The most significant difference was the combination of the four autoantibodies as a panel in addition to the free/total PSA ratio. This combination had the highest area under the curve (AUC)– 0.916 in ROC analysis.

Conclusions

Our results suggest that this autoantibody panel along with PSA and free PSA have potential to segregate patients without cancer from those with prostate cancer with higher sensitivity and specificity than PSA alone.

Assessing Statewide All-Cause Future One-Year Mortality: Prospective Study With Implications for Quality of Life, Resource Utilization, and Medical Futility

Background

For many elderly patients, a disproportionate amount of health care resources and expenditures is spent during the last year of life, despite the discomfort and reduced quality of life associated with many aggressive medical approaches. However, few prognostic tools have focused on predicting all-cause 1-year mortality among elderly patients at a statewide level, an issue that has implications for improving quality of life while distributing scarce resources fairly.

Objective

Using data from a statewide elderly population (aged ≥65 years), we sought to prospectively validate an algorithm to identify patients at risk for dying in the next year for the purpose of minimizing decision uncertainty, improving quality of life, and reducing futile treatment.

Methods

Analysis was performed using electronic medical records from the Health Information Exchange in the state of Maine, which covered records of nearly 95% of the statewide population. The model was developed from 125,896 patients aged at least 65 years who were discharged from any care facility in the Health Information Exchange network from September 5, 2013, to September 4, 2015. Validation was conducted using 153,199 patients with same inclusion and exclusion criteria from September 5, 2014, to September 4, 2016. Patients were stratified into risk groups. The association between all-cause 1-year mortality and risk factors was screened by chi-squared test and manually reviewed by 2 clinicians. We calculated risk scores for individual patients using a gradient tree-based boost algorithm, which measured the probability of mortality within the next year based on the preceding 1-year clinical profile.

Results

The development sample included 125,896 patients (72,572 women, 57.64%; mean 74.2 [SD 7.7] years). The final validation cohort included 153,199 patients (88,177 women, 57.56%; mean 74.3 [SD 7.8] years). The c-statistic for discrimination was 0.96 (95% CI 0.93-0.98) in the development group and 0.91 (95% CI 0.90-0.94) in the validation cohort. The mortality was 0.99% in the low-risk group, 16.75% in the intermediate-risk group, and 72.12% in the high-risk group. A total of 99 independent risk factors (n=99) for mortality were identified (reported as odds ratios; 95% CI). Age was on the top of list (1.41; 1.06-1.48); congestive heart failure (20.90; 15.41-28.08) and different tumor sites were also recognized as driving risk factors, such as cancer of the ovaries (14.42; 2.24-53.04), colon (14.07; 10.08-19.08), and stomach (13.64; 3.26-86.57). Disparities were also found in patients’ social determinants like respiratory hazard index (1.24; 0.92-1.40) and unemployment rate (1.18; 0.98-1.24). Among high-risk patients who expired in our dataset, cerebrovascular accident, amputation, and type 1 diabetes were the top 3 diseases in terms of average cost in the last year of life.

Conclusions

Our study prospectively validated an accurate 1-year risk prediction model and stratification for the elderly population (≥65 years) at risk of mortality with statewide electronic medical record datasets. It should be a valuable adjunct for helping patients to make better quality-of-life choices and alerting care givers to target high-risk elderly for appropriate care and discussions, thus cutting back on futile treatment.

Web-based Real-Time Case Finding for the Population Health Management of Patients With Diabetes Mellitus: A Prospective Validation of the Natural Language Processing-Based Algorithm With Statewide Electronic Medical Records

Background

Diabetes case finding based on structured medical records does not fully identify diabetic patients whose medical histories related to diabetes are available in the form of free text. Manual chart reviews have been used but involve high labor costs and long latency.

Objective

This study developed and tested a Web-based diabetes case finding algorithm using both structured and unstructured electronic medical records (EMRs).

Methods

This study was based on the health information exchange (HIE) EMR database that covers almost all health facilities in the state of Maine, United States. Using narrative clinical notes, a Web-based natural language processing (NLP) case finding algorithm was retrospectively (July 1, 2012, to June 30, 2013) developed with a random subset of HIE-associated facilities, which was then blind tested with the remaining facilities. The NLP-based algorithm was subsequently integrated into the HIE database and validated prospectively (July 1, 2013, to June 30, 2014).

Results

Of the 935,891 patients in the prospective cohort, 64,168 diabetes cases were identified using diagnosis codes alone. Our NLP-based case finding algorithm prospectively found an additional 5756 uncodified cases (5756/64,168, 8.97% increase) with a positive predictive value of .90. Of the 21,720 diabetic patients identified by both methods, 6616 patients (6616/21,720, 30.46%) were identified by the NLP-based algorithm before a diabetes diagnosis was noted in the structured EMR (mean time difference = 48 days).

Conclusions

The online NLP algorithm was effective in identifying uncodified diabetes cases in real time, leading to a significant improvement in diabetes case finding. The successful integration of the NLP-based case finding algorithm into the Maine HIE database indicates a strong potential for application of this novel method to achieve a more complete ascertainment of diagnoses of diabetes mellitus.

A Novel Truncated Form of Serum Amyloid A in Kawasaki Disease

Background

Kawasaki disease (KD) is an acute vasculitis in children that can cause coronary artery abnormalities. Its diagnosis is challenging, and many cytokines, chemokines, acute phase reactants, and growth factors have failed evaluation as specific biomarkers to distinguish KD from other febrile illnesses. We performed protein profiling, comparing plasma from children with KD with febrile control (FC) subjects to determine if there were specific proteins or peptides that could distinguish the two clinical states.

Materials and Methods

Plasma from three independent cohorts from the blood of 68 KD and 61 FC subjects was fractionated by anion exchange chromatography, followed by surface-enhanced laser desorption ionization (SELDI) mass spectrometry of the fractions. The mass spectra of KD and FC plasma samples were analyzed for peaks that were statistically significantly different.

Results

A mass spectrometry peak with a mass of 7,860 Da had high intensity in acute KD subjects compared to subacute KD (p = 0.0003) and FC (p = 7.9 x 10⁻¹⁰) subjects. We identified this peak as a novel truncated form of serum amyloid A with N-terminal at Lys-34 of the circulating form and validated its identity using a hybrid mass spectrum immunoassay technique. The truncated form of serum amyloid A was present in plasma of KD subjects when blood was collected in tubes containing protease inhibitors. This peak disappeared when the patients were examined after their symptoms resolved. Intensities of this peptide did not correlate with KD-associated laboratory values or with other mass spectrum peaks from the plasma of these KD subjects.

Conclusions

Using SELDI mass spectrometry, we have discovered a novel truncated form of serum amyloid A that is elevated in the plasma of KD when compared with FC subjects. Future studies will evaluate its relevance as a diagnostic biomarker and its potential role in the pathophysiology of KD.

Investigation of maternal environmental exposures in association with self-reported preterm birth

Identification of maternal environmental factors influencing preterm birth risks is important to understand the reasons for the increase in prematurity since 1990. Here, we utilized a health survey, the US National Health and Nutrition Examination Survey (NHANES) to search for personal environmental factors associated with preterm birth. 201 urine and blood markers of environmental factors, such as allergens, pollutants, and nutrients were assayed in mothers (range of N: 49-724) who answered questions about any children born preterm (delivery <37 weeks). We screened each of the 201 factors for association with any child born preterm adjusting by age, race/ethnicity, education, and household income. We attempted to verify the top finding, urinary bisphenol A, in an independent study of pregnant women attending Lucile Packard Children's Hospital. We conclude that the association between maternal urinary levels of bisphenol A and preterm birth should be evaluated in a larger epidemiological investigation.

Plasma biomarkers in a mouse model of preterm labor

Preterm labor (PTL) is frequently associated with inflammation. We hypothesized that biomarkers during pregnancy can identify pregnancies most at risk for development of PTL. An inflammation-induced mouse model of PTL was used. Surface-enhanced laser desorption/ionization time-of-flight mass spectrometry was used to analyze and compare the plasma protein (PP) profile between CD-1 mice injected intrauterine with either lipopolysaccharide (LPS) or PBS on d 14.5 of gestation. The median differences of normalized PP peaks between the two groups were determined using the Mann-Whitney U test and the false discovery rate. In a second series of experiments, both groups of mice were injected with a lower dose of LPS. A total of 1665 peaks were detected. Thirty peaks were highly differentially expressed (p < 0.0001) between the groups. Two 11 kDa protein peaks were identified by MALDI-TOF/TOF-MS and confirmed to be mouse serum amyloid A (SAA) 1 and 2. Plasma SAA2 levels were increased in LPS-treated animals compared with controls and in LPS-treated animals that delivered preterm vs. those that delivered at term. SAA2 has the potential to be a plasma biomarker that can identify pregnancies at risk for development of PTL.

FDR made easy in differential feature discovery and correlation analyses

SUMMARY

Rapid progress in technology, particularly in high-throughput biology, allows the analysis of thousands of genes or proteins simultaneously, where the multiple comparison problems occurs. Global false discovery rate (gFDR) analysis statistically controls this error, computing the ratio of the number of false positives over the total number of rejections. Local FDR (lFDR) method can associate the corrected significance measure with each hypothesis testing for its feature-by-feature interpretation. Given the large feature number and sample size in any genomics or proteomics analysis, FDR computation, albeit critical, is both beyond the regular biologists' specialty and computationally expensive, easily exceeding the capacity of desktop computers. To overcome this digital divide, a web portal has been developed that provides bench-side biologists easy access to the server-side computing capabilities to analyze for FDR, differential expressed genes or proteins, and for the correlation between molecular data and clinical measurements.

AVAILABILITY

(http://translationalmedicine.stanford.edu/Mass-Conductor/FDR.html).

GO-Diff: mining functional differentiation between EST-based transcriptomes

Background

Large-scale sequencing efforts produced millions of Expressed Sequence Tags (ESTs) collectively representing differentiated biochemical and functional states. Analysis of these EST libraries reveals differential gene expressions, and therefore EST data sets constitute valuable resources for comparative transcriptomics. To translate differentially expressed genes into a better understanding of the underlying biological phenomena, existing microarray analysis approaches usually involve the integration of gene expression with Gene Ontology (GO) databases to derive comparable functional profiles. However, methods are not available yet to process EST-derived transcription maps to enable GO-based global functional profiling for comparative transcriptomics in a high throughput manner.

Results

Here we present GO-Diff, a GO-based functional profiling approach towards high throughput EST-based gene expression analysis and comparative transcriptomics. Utilizing holistic gene expression information, the software converts EST frequencies into EST Coverage Ratios of GO Terms. The ratios are then tested for statistical significances to uncover differentially represented GO terms between the compared transcriptomes, and functional differences are thus inferred. We demonstrated the validity and the utility of this software by identifying differentially represented GO terms in three application cases: intra-species comparison; meta-analysis to test a specific hypothesis; inter-species comparison. GO-Diff findings were consistent with previous knowledge and provided new clues for further discoveries. A comprehensive test on the GO-Diff results using series of comparisons between EST libraries of human and mouse tissues showed acceptable levels of consistency: 61% for human-human; 69% for mouse-mouse; 47% for human-mouse.

Conclusion

GO-Diff is the first software integrating EST profiles with GO knowledge databases to mine functional differentiation between biological systems, e.g. tissues of the same species or the same tissue cross species. With rapid accumulation of EST resources in the public domain and expanding sequencing effort in individual laboratories, GO-Diff is useful as a screening tool before undertaking serious expression studies.

Multiclass cancer classification and biomarker discovery using GA-based algorithms

MOTIVATION

The development of microarray-based high-throughput gene profiling has led to the hope that this technology could provide an efficient and accurate means of diagnosing and classifying tumors, as well as predicting prognoses and effective treatments. However, the large amount of data generated by microarrays requires effective reduction of discriminant gene features into reliable sets of tumor biomarkers for such multiclass tumor discrimination. The availability of reliable sets of biomarkers, especially serum biomarkers, should have a major impact on our understanding and treatment of cancer.

RESULTS

We have combined genetic algorithm (GA) and all paired (AP) support vector machine (SVM) methods for multiclass cancer categorization. Predictive features can be automatically determined through iterative GA/SVM, leading to very compact sets of non-redundant cancer-relevant genes with the best classification performance reported to date. Interestingly, these different classifier sets harbor only modest overlapping gene features but have similar levels of accuracy in leave-one-out cross-validations (LOOCV). Further characterization of these optimal tumor discriminant features, including the use of nearest shrunken centroids (NSC), analysis of annotations and literature text mining, reveals previously unappreciated tumor subclasses and a series of genes that could be used as cancer biomarkers. With this approach, we believe that microarray-based multiclass molecular analysis can be an effective tool for cancer biomarker discovery and subsequent molecular cancer diagnosis.

Molecular classification of cancer types from microarray data using the combination of genetic algorithms and support vector machines

Simultaneous multiclass classification of tumor types is essential for future clinical implementations of microarray-based cancer diagnosis. In this study, we have combined genetic algorithms (GAs) and all paired support vector machines (SVMs) for multiclass cancer identification. The predictive features have been selected through iterative SVMs/GAs, and recursive feature elimination post-processing steps, leading to a very compact cancer-related predictive gene set. Leave-one-out cross-validations yielded accuracies of 87.93% for the eight-class and 85.19% for the fourteen-class cancer classifications, outperforming the results derived from previously published methods.

A comparative analysis of HGSC and Celera human genome assemblies and gene sets

MOTIVATION

Since the simultaneous publication of the human genome assembly by the International Human Genome Sequencing Consortium (HGSC) and Celera Genomics, several comparisons have been made of various aspects of these two assemblies. In this work, we set out to provide a more comprehensive comparative analysis of the two assemblies and their associated gene sets.

RESULTS

The local sequence content for both draft genome assemblies has been similar since the early releases, however it took a year for the quality of the Celera assembly to approach that of HGSC, suggesting an advantage of HGSC's hierarchical shotgun (HS) sequencing strategy over Celera's whole genome shotgun (WGS) approach. While similar numbers of ab initio predicted genes can be derived from both assemblies, Celera's Otto approach consistently generated larger, more varied gene sets than the Ensembl gene build system. The presence of a non-overlapping gene set has persisted with successive data releases from both groups. Since most of the unique genes from either genome assembly could be mapped back to the other assembly, we conclude that the gene set discrepancies do not reflect differences in local sequence content but rather in the assemblies and especially the different gene-prediction methodologies.

PRC17, a novel oncogene encoding a Rab GTPase-activating protein, is amplified in prostate cancer

We used cDNA-based genomic microarrays to examine DNA copy number changes in a panel of prostate tumors and found a previously undescribed amplicon on chromosome 17 containing a novel overexpressed gene that we termed prostate cancer gene 17 (PRC17). When overexpressed in 3T3 mouse fibroblast cells, PRC17 induced growth in low serum, loss of contact inhibition, and tumor formation in nude mice. The PRC17 gene product contains a GTPase-activating protein (GAP) catalytic core motif found in various Rab/Ypt GAPs, including RN-Tre. Similar to RN-Tre, we found that PRC17 protein interacts directly with Rab5 and stimulates its GTP hydrolysis. Point mutations that alter conserved amino acid residues within the PRC17 GAP domain abolished its transforming abilities, suggesting that GAP activity is essential for its oncogenic function. Whereas PRC17 is amplified in 15% of prostate cancers, it is highly overexpressed in approximately one-half of metastatic prostate tumors. The potent oncogenic activity of PRC17 is likely to influence the tumorigenic phenotype of these prostate cancers.

Comparative analysis of human genome assemblies reveals genome-level differences

Previous comparative analysis has revealed a significant disparity between the predicted gene sets produced by the International Human Genome Sequencing Consortium (HGSC) and Celera Genomics. To determine whether the source of this discrepancy was due to underlying differences in the genomic sequences or different gene prediction methodologies, we analyzed both genome assemblies in parallel. Using the GENSCAN gene prediction algorithm, we generated predicted transcriptomes that could be directly compared. BLAST-based comparisons revealed a 20-30% difference between the transcriptomes. Further differences between the two genomes were revealed with protein domain PFAM analyses. These results suggest that fundamental differences between the two genome assemblies are likely responsible for a significant portion of the discrepancy between the transcript sets predicted by the two groups.

DIAN: a novel algorithm for genome ontological classification

Faced with the determination of many completely sequenced genomes, computational biology is now faced with the challenge of interpreting the significance of these data sets. A multiplicity of data-related problems impedes this goal: Biological annotations associated with raw data are often not normalized, and the data themselves are often poorly interrelated and their interpretation unclear. All of these problems make interpretation of genomic databases increasingly difficult. With the current explosion of sequences now available from the human genome as well as from model organisms, the importance of sorting this vast amount of conceptually unstructured source data into a limited universe of genes, proteins, functions, structures, and pathways has become a bottleneck for the field. To address this problem, we have developed a method of interrelating data sources by applying a novel method of associating biological objects to ontologies. We have developed an intelligent knowledge-based algorithm, to support biological knowledge mapping, and, in particular, to facilitate the interpretation of genomic data. In this respect, the method makes it possible to inventory genomes by collapsing multiple types of annotations and normalizing them to various ontologies. By relying on a conceptual view of the genome, researchers can now easily navigate the human genome in a biologically intuitive, scientifically accurate manner.

Induction of triploid Citrus plants from endosperm calli in vitro

Triploid hybrid Citrus plants were regenerated by somatic embryogenesis in vitro from endosperm derived calli. A sequence of media formulations was used to induce and support proliferation of primary callus from endosperm, to induce embryogenesis from primary callus, and to allow embryo development leading to viable plantlets. Calli were induced from cellular endosperm of Citrus sinensis (sweet orange), C. Xparadisi (grapefruit), and C. grandis (pummelo) excised 12-14 weeks post-anthesis. Induction of embryogenesis from sweet orange and pummelo primary calli required gibberellic acid and double mineral nutrient concentrations. Embryogenesis was not induced from grapefruit calli in these experiments. Only sweet orange embryos developed sufficiently to allow plant regeneration. Triploid axillary buds were minigrafted onto etiolated diploid rootstock seedlings in vitro in order to transfer triploid regenerants to soil and the external environment. Triploidy (2n = 3x = 27) was observed consistently in all phases of regeneration and in recovered plants. These results demonstrate that triploid hybrid plant recovery from Citrus endosperm can overcome barriers to sexual hybridization resulting from apomixis.