Repurposing non-pharmacological interventions for Alzheimer's disease through link prediction on biomedical literature.

Non-pharmaceutical interventions (NPI) have great potential to improve cognitive function but limited investigation to discover NPI repurposing for Alzheimer's Disease (AD). This is the first study to develop an innovative framework to extract and represent NPI information from biomedical literature in a knowledge graph (KG), and train link prediction models to repurpose novel NPIs for AD prevention. We constructed a comprehensive KG, called ADInt, by extracting NPI information from biomedical literature. We used the previously-created SuppKG and NPI lexicon to identify NPI entities. Four KG embedding models (i.e., TransE, RotatE, DistMult and ComplEX) and two novel graph convolutional network models (i.e., R-GCN and CompGCN) were trained and compared to learn the representation of ADInt. Models were evaluated and compared on two test sets (time slice and clinical trial ground truth) and the best performing model was used to predict novel NPIs for AD. Discovery patterns were applied to generate mechanistic pathways for high scoring candidates. The ADInt has 162,212 nodes and 1,017,284 edges. R-GCN performed best in time slice (MR = 5.2054, Hits@10 = 0.8496) and clinical trial ground truth (MR = 3.4996, Hits@10 = 0.9192) test sets. After evaluation by domain experts, 10 novel dietary supplements and 10 complementary and integrative health were proposed from the score table calculated by R-GCN. Among proposed novel NPIs, we found plausible mechanistic pathways for photodynamic therapy and Choerospondias axillaris to prevent AD, and validated psychotherapy and manual therapy techniques using real-world data analysis. The proposed framework shows potential for discovering new NPIs for AD prevention and understanding their mechanistic pathways.

Repurposing Non-pharmacological Interventions for Alzheimer's Diseases through Link Prediction on Biomedical Literature.

Recently, computational drug repurposing has emerged as a promising method for identifying new pharmaceutical interventions (PI) for Alzheimer's Disease (AD). Non-pharmaceutical interventions (NPI), such as Vitamin E and Music therapy, have great potential to improve cognitive function and slow the progression of AD, but have largely been unexplored. This study predicts novel NPIs for AD through link prediction on our developed biomedical knowledge graph. We constructed a comprehensive knowledge graph containing AD concepts and various potential interventions, called ADInt, by integrating a dietary supplement domain knowledge graph, SuppKG, with semantic relations from SemMedDB database. Four knowledge graph embedding models (TransE, RotatE, DistMult and ComplEX) and two graph convolutional network models (R-GCN and CompGCN) were compared to learn the representation of ADInt. R-GCN outperformed other models by evaluating on the time slice test set and the clinical trial test set and was used to generate the score tables of the link prediction task. Discovery patterns were applied to generate mechanism pathways for high scoring triples. Our ADInt had 162,213 nodes and 1,017,319 edges. The graph convolutional network model, R-GCN, performed best in both the Time Slicing test set (MR = 7.099, MRR = 0.5007, Hits@1 = 0.4112, Hits@3 = 0.5058, Hits@10 = 0.6804) and the Clinical Trials test set (MR = 1.731, MRR = 0.8582, Hits@1 = 0.7906, Hits@3 = 0.9033, Hits@10 = 0.9848). Among high scoring triples in the link prediction results, we found the plausible mechanism pathways of (Photodynamic therapy, PREVENTS, Alzheimer's Disease) and (Choerospondias axillaris, PREVENTS, Alzheimer's Disease) by discovery patterns and discussed them further. In conclusion, we presented a novel methodology to extend an existing knowledge graph and discover NPIs (dietary supplements (DS) and complementary and integrative health (CIH)) for AD. We used discovery patterns to find mechanisms for predicted triples to solve the poor interpretability of artificial neural networks. Our method can potentially be applied to other clinical problems, such as discovering drug adverse reactions and drug-drug interactions.

Effect of iron supplementation from neonatal period on the iron status of6-month-old infants at-risk for early iron deficiency: a randomized interventional trial.

OBJECTIVE: To investigate the effects of iron supplementation from the second day after birth on 6-month hemoglobin (Hb), serum ferritin and motor development in infants at risk of early iron deficiency. STUDY DESIGN: Term (37-41 weeks) infants of anemic (Hb ≤ 100 g L-1; N = 100) and non-anemic (Hb > 100 g L-1; N = 100) mothers were randomized to daily iron supplementation at a dose of 2 mg kg-1 from 36 h of age (N = 50, each of anemic and non-anemic mothers) or no iron-supplementation (N = 50 each of anemic and non-anemic mothers). Hb, serum ferritin and motor development at 6 months were compared in the two groups. RESULTS: Iron-supplemented infants had higher Hb (103.7 ± 9.3 g L-1 versus 97.0 ± 9.4 g L-1, p < .0001) and serum ferritin (133.93 ± 52.41 ng mL-1 versus 78.09 ± 42.03 ng mL-1, p < .001) concentrations, compared with the no iron-supplementation group. Their motor development was closer to age-appropriate norms than the no iron-supplementation group (5.83 ± 0.69 versus 5.18 ± 1.35, p < .01). CONCLUSION: Early Iron supplementation is effective for improving iron status and motor development at 6 months in infants at risk for early iron deficiency.

A Naive Bayes machine learning approach to risk prediction using censored, time-to-event data.

Predicting an individual's risk of experiencing a future clinical outcome is a statistical task with important consequences for both practicing clinicians and public health experts. Modern observational databases such as electronic health records provide an alternative to the longitudinal cohort studies traditionally used to construct risk models, bringing with them both opportunities and challenges. Large sample sizes and detailed covariate histories enable the use of sophisticated machine learning techniques to uncover complex associations and interactions, but observational databases are often 'messy', with high levels of missing data and incomplete patient follow-up. In this paper, we propose an adaptation of the well-known Naive Bayes machine learning approach to time-to-event outcomes subject to censoring. We compare the predictive performance of our method with the Cox proportional hazards model which is commonly used for risk prediction in healthcare populations, and illustrate its application to prediction of cardiovascular risk using an electronic health record dataset from a large Midwest integrated healthcare system.

Fortification of staple cereal flours with iron and other micronutrients: cost implications of following World Health Organization-endorsed recommendations.

BACKGROUND: Wheat and maize flours are widely used delivery vehicles for mass fortification. In lower-income countries, most, if not all, national-level cereal flour fortification programs routinely fortify with iron; however, cofortification with other micronutrients is common. Little information is available on the cost implications programs face when considering current fortification practices versus what the World Health Organization (WHO) interim consensus statement recommends. OBJECTIVE: The objectives of the present paper are to provide information on the costs of adding different chemical forms of iron and/or other micronutrients to premix formulations, and to discuss some of the issues that should be considered regarding which micronutrients to include in the premix. METHODS: Nine countries in Latin America, Africa, and Asia (three countries per region) that currently cofortify with multiple micronutrients including iron were selected based on low (< 75 g/day), medium (75 to 149 g/day), and high (> or = 150 g/day) mean population flour consumption levels. Premix costs per metric ton of flour produced were estimated for improving iron formulations and for following WHO recommendations for other micronutrients. RESULTS: For the selected programs to maintain current premix formulations and improve iron compounds, premix costs would increase by between 155% and 343% when the iron compound was switched from electrolytic iron to sodium iron ethylenediaminetetraacetate (NaFeEDTA), by 6% to 50% when it was switched from electrolytic iron to ferrous sulfate, and by 4% to 13 when iron addition rates were adjusted without switching the compound. To meet WHO recommendations for other micronutrients, premix costs would increase the most when the amounts of vitamins B12 and A were increased. CONCLUSIONS: For programs that currently cofortify with iron and additional micronutrients, the quality of the iron fortificant should not be overlooked simply to be able to afford to add more micronutrients to the premix. Micronutrients should be selected according to population needs, costs, and potential beneficial synergistic reactions the added micronutrients may have.

Demonstrating zinc and iron bioavailability from intrinsically labeled microencapsulated ferrous fumarate and zinc gluconate Sprinkles in young children.

Nutrient-nutrient interactions are an important consideration for any multiple-micronutrient formulation, including Sprinkles, a home-fortification strategy to control anemia. The objectives of this randomized controlled trial were as follows: 1) to compare the absorption of zinc at 2 doses given as Sprinkles; and 2) to examine the effect of zinc and ascorbic acid (AA) on iron absorption from Sprinkles. Seventy-five children aged 12-24 mo were randomly assigned to the following groups: 1) 5 mg of labeled zinc (67Zn) with 50 mg AA (LoZn group); b) 10 mg of labeled zinc (67Zn) with 50 mg AA (HiZn group); or 3) 5 mg zinc with no AA (control). All groups contained 30 mg of labeled iron (57Fe). Intravenous infusions labeled with 70Zn (LoZn and HiZn groups) and 58Fe (control) were administered. Blood was drawn at baseline, 48 h and 14 d later. The percentage of zinc absorbed did not differ between LoZn (geometric mean = 6.4%; min-max: 1.7-14.6) and HiZn (geometric mean = 7.5%; min-max: 3.3-18.0) groups. However, total zinc absorbed was significantly different between the LoZn (geometric mean = 0.31 mg; min-max: 0.08-0.73) and HiZn (geometric mean = 0.82 mg; min-max: 0.33-1.82) groups (P = 0.0004). Geometric mean percentage iron absorption values did not differ between the LoZn (5.9%; min-max: 0.8-21) and HiZn (4.4%; min-max: 0.6-12.3) groups and between the LoZn and control groups (5.0%; min-max: 1.4-24). We conclude that zinc in the form of Sprinkles has a low bioavailability, yet provides adequate amounts of absorbed zinc in young children, and that there is no effect of zinc or AA on iron absorption from the given formulations of Sprinkles.