Deep learning of immune cell differentiation.

Although we know many sequence-specific transcription factors (TFs), how the DNA sequence of cis-regulatory elements is decoded and orchestrated on the genome scale to determine immune cell differentiation is beyond our grasp. Leveraging a granular atlas of chromatin accessibility across 81 immune cell types, we asked if a convolutional neural network (CNN) could learn to infer cell type-specific chromatin accessibility solely from regulatory DNA sequences. With a tailored architecture and an ensemble approach to CNN parameter interpretation, we show that our trained network ("AI-TAC") does so by rediscovering ab initio the binding motifs for known regulators and some unknown ones. Motifs whose importance is learned virtually as functionally important overlap strikingly well with positions determined by chromatin immunoprecipitation for several TFs. AI-TAC establishes a hierarchy of TFs and their interactions that drives lineage specification and also identifies stage-specific interactions, like Pax5/Ebf1 vs. Pax5/Prdm1, or the role of different NF-κB dimers in different cell types. AI-TAC assigns Spi1/Cebp and Pax5/Ebf1 as the drivers necessary for myeloid and B lineage fates, respectively, but no factors seemed as dominantly required for T cell differentiation, which may represent a fall-back pathway. Mouse-trained AI-TAC can parse human DNA, revealing a strikingly similar ranking of influential TFs and providing additional support that AI-TAC is a generalizable regulatory sequence decoder. Thus, deep learning can reveal the regulatory syntax predictive of the full differentiative complexity of the immune system.

Partner expressed emotion and diabetes management among spouses living with Type 2 diabetes.

INTRODUCTION: Expressed emotion has been consistently shown to be a significant predictor of relapse and poor disease management across numerous physical and mental health conditions, however very little research has been conducted on its relationship to the management practices of individuals living with Type 2 diabetes. This study examines the relationship between expressed emotion (EE) and diabetes management among couples where 1 spouse has Type 2 diabetes. METHODS: The authors surveyed 106 couples where 1 partner was diagnosed with Type 2 diabetes. Partners without diabetes completed questionnaires about their level of criticism, emotional involvement, and warmth toward their partners. Partners with diabetes completed questionnaires on diabetes control, diabetes management practices and attitude toward their diabetes. RESULTS: The authors found that, individuals living with diabetes who had partners with high EE reported significantly poorer diabetes management in all areas (diet, physical activity, and attitude toward diabetes). Diabetes management was found to mediate the relationship between EE and diabetes control. Results suggest that partners with high EE may have a significant influence on diabetes management practices in their partner. DISCUSSION: These findings highlight the important role couple interactions may play in diabetes management. Findings also emphasize the potential benefit of conceptualizing diabetes management from a systems/relational perspective. In addition, greater consideration should be given to using family-based approaches for diabetes management and treatment among coupled individuals living with Type 2 diabetes. (PsycINFO Database Record

Paternal depression during pregnancy and postpartum: An international Delphi study.

BACKGROUND: Fathers are at risk for depression during a mother's gestation and postpartum. Assessment, detection, and treatment are hampered by the lack of consensus on this issue. The purpose of this study was to reach expert consensus through the Delphi method on the defining factors of depression in peripartum fathers. METHODS: Purposive sampling resulted in the surveying of 14 international expert panelists. The study used a modified Delphi approach in which experts participated in two rounds of open-ended and scale questionnaires, followed by two rounds of opportunities to adjust their responses and/or comment on evolving data until consensus was achieved. RESULTS: Experts responded to 10 questions on terminology, diagnostics, symptomology, risk/protective factors, biological factors, assessment tools/protocol, cost implications, and key stakeholders. Of these 10 questions presented for discussion, the analysis resulted in 197 coded themes. Consensus was met for 119 of the 197 coded responses (60.41%). LIMITATIONS: Diversity of opinion within this Delphi Study was excluded for the sake of consensus. Regression to the mean may have occurred after continuous surveying and when evolving results were shared with panelists. Critics of Delphi methodologies have pointed to the issue of small expert samples typically used and the subjectivity of "expert." CONCLUSION: Consensus identified diagnostic criteria and symptomology that differentiates the paternal experience of peripartum depression. Experts indicated the importance of a father's social context, biological risk factors, limitations of current assessment tools, key stakeholders, and potential financial costs. Stakeholders on this issue would benefit from translating consensus into assessment and treatment.

Exploring family factors and sexual behaviors in a group of Black and Hispanic adolescent males.

OBJECTIVE: To explore family factors that influence sexual behaviors among a group of Black and Hispanic adolescent males. METHODS: One hundred seventy-eight subjects were selected from a convenience sample of 431 adolescents who participated in a 1996 Youth Survey conducted in San Bernardino County, California. RESULTS: Having a sibling who was a teen parent was significantly associated with engaging in sexual intercourse among Hispanics. Among Blacks, having a father who was a teen dad was significantly associated with sexual intercourse. CONCLUSIONS: Findings showed relationships between family characteristics and sexual attitudes and behaviors among participants.