Gene x abstinence effects on drug cue reactivity in addiction: multimodal evidence.

Functional polymorphisms in the dopamine transporter gene (DAT1 or SLC6A3) modulate responsiveness to salient stimuli, such that carriers of one 9R-allele of DAT1 (compared with homozygote carriers of the 10R-allele) show heightened reactivity to drug-related reinforcement in addiction. Here, using multimodal neuroimaging and behavioral dependent variables in 73 human cocaine-addicted individuals and 47 healthy controls, we hypothesized and found that cocaine-addicted carriers of a 9R-allele exhibited higher responses to drug cues, but only among individuals who had used cocaine within 72 h of the study as verified by positive cocaine urine screens (a state characterized by intense craving). Importantly, this responsiveness to drug cues was reliably preserved across multimodal imaging and behavioral probes: psychophysiological event-related potentials, self-report, simulated cocaine choice, and fMRI. Because drug cues contribute to relapse, our results identify the DAT1R 9R-allele as a vulnerability allele for relapse especially during early abstinence (e.g., detoxification).

An automatic integrative method for learning interpretable communities of biological pathways.

Although knowledge of biological pathways is essential for interpreting results from computational biology studies, the growing number of pathway databases complicates efforts to efficiently perform pathway analysis due to high redundancies among pathways from different databases, and inconsistencies in how pathways are created and named. We introduce the PAthway Communities (PAC) framework, which reconciles pathways from different databases and reduces pathway redundancy by revealing informative groups with distinct biological functions. Uniquely applying the Louvain community detection algorithm to a network of 4847 pathways from KEGG, REACTOME and Gene Ontology databases, we identify 35 distinct and automatically annotated communities of pathways and show that they are consistent with expert-curated pathway categories. Further, we demonstrate that our pathway community network can be queried with new gene sets to provide biological context in terms of related pathways and communities. Our approach, combined with an interpretable web tool we provide, will help computational biologists more efficiently contextualize and interpret their biological findings.

Efficient and Explainable Risk Assessments for Imminent Dementia in an Aging Cohort Study.

As the aging US population grows, scalable approaches are needed to identify individuals at risk for dementia. Common prediction tools have limited predictive value, involve expensive neuroimaging, or require extensive and repeated cognitive testing. None of these approaches scale to the sizable aging population who do not receive routine clinical assessments. Our study seeks a tractable and widely administrable set of metrics that can accurately predict imminent (i.e., within three years) dementia onset. To this end, we develop and apply a machine learning (ML) model to an aging cohort study with an extensive set of longitudinal clinical variables to highlight at-risk individuals with better accuracy than standard rudimentary approaches. Next, we reduce the burden needed to achieve accurate risk assessments for those deemed at risk by (1) predicting when consecutive clinical visits may be unnecessary, and (2) selecting a subset of highly predictive cognitive tests. Finally, we demonstrate that our method successfully provides individualized prediction explanations that retain non-linear feature effects present in the data. Our final model, which uses only four cognitive tests (less than 20 minutes to administer) collected in a single visit, affords predictive performance comparable to a standard 100-minute neuropsychological battery and personalized risk explanations. Our approach shows the potential for an efficient tool for screening and explaining dementia risk in the general aging population.

Unified AI framework to uncover deep interrelationships between gene expression and Alzheimer's disease neuropathologies.

Deep neural networks (DNNs) capture complex relationships among variables, however, because they require copious samples, their potential has yet to be fully tapped for understanding relationships between gene expression and human phenotypes. Here we introduce an analysis framework, namely MD-AD (Multi-task Deep learning for Alzheimer's Disease neuropathology), which leverages an unexpected synergy between DNNs and multi-cohort settings. In these settings, true joint analysis can be stymied using conventional statistical methods, which require "harmonized" phenotypes and tend to capture cohort-level variations, obscuring subtler true disease signals. Instead, MD-AD incorporates related phenotypes sparsely measured across cohorts, and learns interactions between genes and phenotypes not discovered using linear models, identifying subtler signals than cohort-level variations which can be uniquely recapitulated in animal models and across tissues. We show that MD-AD exploits sex-specific relationships between microglial immune response and neuropathology, providing a nuanced context for the association between inflammatory genes and Alzheimer's Disease.

Effects of an opioid (proenkephalin) polymorphism on neural response to errors in health and cocaine use disorder.

Chronic exposure to drugs of abuse perturbs the endogenous opioid system, which plays a critical role in the development and maintenance of addictive disorders. Opioid genetics may therefore play an important modulatory role in the expression of substance use disorders, but these genes have not been extensively characterized, especially in humans. In the current imaging genetics study, we investigated a single nucleotide polymorphism (SNP) of the protein-coding proenkephalin gene (PENK: rs2609997, recently shown to be associated with cannabis dependence) in 55 individuals with cocaine use disorder and 37 healthy controls. Analyses tested for PENK associations with fMRI response to error (during a classical color-word Stroop task) and gray matter volume (voxel-based morphometry) as a function of Diagnosis (cocaine, control). Results revealed whole-brain Diagnosis×PENK interactions on the neural response to errors (fMRI error>correct contrast) in the right putamen, left rostral anterior cingulate cortex/medial orbitofrontal cortex, and right inferior frontal gyrus; there was also a significant Diagnosis×PENK interaction on right inferior frontal gyrus gray matter volume. These interactions were driven by differences between individuals with cocaine use disorders and controls that were accentuated in individuals carrying the higher-risk PENK C-allele. Taken together, the PENK polymorphism-and potentially opioid neurotransmission more generally-modulates functioning and structural integrity of brain regions previously implicated in error-related processing. PENK could potentially render a subgroup of individuals with cocaine use disorder (i.e., C-allele carriers) more sensitive to mistakes or other related challenges; in future studies, these results could contribute to the development of individualized genetics-informed treatments.

Monoamine polygenic liability in health and cocaine dependence: imaging genetics study of aversive processing and associations with depression symptomatology.

BACKGROUND: Gene polymorphisms that affect serotonin signaling modulate reactivity to salient stimuli and risk for emotional disturbances. Here, we hypothesized that these serotonin genes, which have been primarily explored in depressive disorders, could also have important implications for drug addiction, with the potential to reveal important insights into drug symptomatology, severity, and/or possible sequelae such as dysphoria. METHODS: Using an imaging genetics approach, the current study tested in 62 cocaine abusers and 57 healthy controls the separate and combined effects of variations in the serotonin transporter (5-HTTLPR) and monoamine oxidase A (MAOA) genes on processing of aversive information. Reactivity to standardized unpleasant images was indexed by a psychophysiological marker of stimulus salience (i.e., the late positive potential (LPP) component of the event-related potential) during passive picture viewing. Depressive symptomatology was assessed with the Beck Depression Inventory (BDI). RESULTS: Results showed that, independent of diagnosis, the highest unpleasant LPPs emerged in individuals with MAOA-Low and at least one 'Short' allele of 5-HTTLPR. Uniquely in the cocaine participants with these two risk variants, higher unpleasant LPPs correlated with higher BDI scores. CONCLUSIONS: Taken together, these results suggest that a multilocus genetic composite of monoamine signaling relates to depression symptomatology through brain function associated with the experience of negative emotions. This research lays the groundwork for future studies that can investigate clinical outcomes and/or pharmacogenetic therapies in drug addiction and potentially other psychopathologies of emotion dysregulation.

Choice to view cocaine images predicts concurrent and prospective drug use in cocaine addiction.

BACKGROUND: Identifying variables that predict drug use in treatment-seeking drug addicted individuals is a crucial research and therapeutic goal. This study tested the hypothesis that choice to view cocaine images is associated with concurrent and prospective drug use in cocaine addiction. METHODS: To establish choice-concurrent drug use associations, 71 cocaine addicted subjects (43 current users and 28 treatment seekers) provided data on (A) choice to view cocaine images and affectively pleasant, unpleasant, and neutral images [collected under explicit contingencies (when choice was made between two fully visible side-by-side images) and under more probabilistic contingencies (when choice was made between pictures hidden under flipped-over cards)]; and (B) past-month cocaine and other drug use. To establish choice-prospective drug use associations, 20 of these treatment-seeking subjects were followed over the next 6 months. RESULTS: Baseline cocaine-related picture choice as measured by both tasks positively correlated with subjects' concurrent cocaine and other drug use as driven by the actively-using subjects. In a subsequent multiple regression analysis, choice to view cocaine images as compared with affectively pleasant images (under probabilistic contingencies) was the only predictor that continued to be significantly associated with drug use. Importantly, this same baseline cocaine>pleasant probabilistic choice also predicted the number of days drugs were used (cocaine, alcohol, and marijuana) over the next 6 months. CONCLUSIONS: Simulated cocaine choice - especially when probabilistic and when compared with other positive reinforcers - may provide a valid laboratory marker of current and future drug use in cocaine addiction.