Many adolescents participate in risky and excessive alcohol consumption behaviors (Crews et al., 2007), especially in European and North American countries. Several studies have identified that such early and risky exposure to alcohol is a potential risk factor that can lead to the development of Alcohol Use Disorder (AUD) later in life (DeWit et al., 2000; Grant et al., 2006; Nixon and McClain, 2010). During adolescence and early adulthood (age 10–24), the human brain undergoes maturation characterized by an increase in white matter (WM) (Lebel and Beaulieu, 2011) and an initial thickening and later thinning of grey matter (GM) regions (Giedd, 2004). Researchers have suggested that excessive alcohol use during this period might disrupt normal brain maturation, causing lifelong effects (Crews et al., 2007; Monti et al., 2005; Chambers et al., 2003). Therefore, understanding how alcohol misuse during adolescence is related to the development of Alcohol Use Disorder (AUD) later in life is crucial to understanding alcohol addiction. Furthermore, uncovering how adolescent alcohol misuse (AAM) is associated with their brain at different stages of adolescent brain development can help to implement a more informed public health policy surrounding alcohol use during this age. Previous studies: Several studies in the last two decades have attempted to uncover how adolescent alcohol misuse (AAM) and their structural brain are related. These are summarised in Table 1. Earlier studies collected data with small sample size of 30–100 subjects and compared specific brain regions (such as the hippocampus or the pre-frontal cortex (pFC)) between adolescent alcohol misusers (AAMs) and mild users or non-users (controls). They used structural features such as regional volume (De Bellis et al., 2000; Nagel et al., 2005; De Bellis et al., 2005), cortical thickness (Squeglia et al., 2012), or white matter tract volumes (McQueeny et al., 2009; Jones and Nagel, 2019). These studies found differences between the groups in regions such as the hippocampus (De Bellis et al., 2000; Nagel et al., 2005), cerebellum (De Bellis et al., 2005), and the frontal cortex (De Bellis et al., 2005). However, these findings are not always consistent across studies (Jones et al., 2018). This inconsistency is also evident from the findings in the last column of Table 1. Another group of studies investigated into whether AAM disrupts the natural developmental trajectory of adolescent brains (Jacobus et al., 2013; Luciana et al., 2013; Pfefferbaum et al., 2018; Jones and Nagel, 2019; Sullivan et al., 2020; Robert et al., 2020). These studies reported that the brains of AAMs showed accelerated GM decline (Luciana et al., 2013; Pfefferbaum et al., 2018; Sullivan et al., 2020) and attenuated WM growth (Luciana et al., 2013; Sullivan et al., 2020) compared to controls. However, brain regions reported were not consistent between these studies either and do not tell a coherent story (Jones et al., 2018) (see Table 1). These differences in findings could be potentially due to the following reasons:

1. Heterogeneous disease with a weak effect size: Alcohol misuse has a heterogeneous expression in the brain (Zahr and Pfefferbaum, 2017). This heterogeneity might be driven by alcohol misuse affecting diverse brain regions in different sub-populations depending on demographic, environmental, or genetic differences (Grant et al., 2015). Furthermore, the effect of alcohol misuse on adolescent brain structure can be weak and hard to detect (especially with the mass-univariate methods used in previous studies). The possibility of several disease subtypes exasperated by the small signal-to-noise ratio can generate incoherent findings regarding which brain regions are affected by alcohol.

2. Higher risk of false-positives: Most previous studies have small sample size that are prone to generate inflated effect size (Button et al., 2013). Furthermore, these studies employ mass-univariate analysis techniques that are vulnerable to multiple comparisons problem (Lindquist and Mejia, 2015) and can produce false-positives if ignored. These factors coupled with the possibility of publication bias to produce positive results (Ioannidis, 2005) can have a high likelihood of generating false-positive findings (Scheel et al., 2021).

3. Several metrics to measure alcohol misuse: There is no consensus on what is the best phenotype to measure AAM. Many studies use binge drinking or heavy episodic drinking as a measure of AAM (Squeglia et al., 2012; Whelan et al., 2014; Jones and Nagel, 2019; Robert et al., 2020), while few others use a combination of binge drinking, frequency of alcohol use, amount of alcohol consumed and the age of onset of alcohol misuse (Squeglia et al., 2015; Pfefferbaum et al., 2018; Kühn et al., 2019; Seo et al., 2019; Sullivan et al., 2020). These differences in analyses could potentially produce different findings.

Multivariate exploratory analysis: Over the last years, data collection drives such as IMAGEN (Mascarell Maričić et al., 2020), NCANDA (Brown et al., 2015), and UK Biobank (Sudlow et al., 2015) made available large-sample multi-site data with ${\displaystyle n>1000}$ that are representative of the general population. This enabled researchers to use multivariate, data-driven, and exploratory analysis tools such as machine learning (ML) to detect effects of alcohol misuse on multiple brain regions (Whelan et al., 2014; Squeglia et al., 2017; Seo et al., 2019; Filippi et al., 2021; Jia et al., 2021; Yip et al., 2022). Such whole-brain multivariate methods are preferable over the previous mass-univariate methods as they have a higher sensitivity to detect true positives (Hebart and Baker, 2018). Furthermore, ML can be easily used for clinical applications such as computer-aided diagnosis, predicting future development of AUD, and future relapse of patients into AUD (Shiraishi et al., 2011).

Due to these advantages, several exploratory studies using ML have been attempted in AUD research (Whelan et al., 2014; Seo et al., 2019; Squeglia et al., 2017). We further extend this line of work by analyzing the newly available longitudinal data from IMAGEN ($n\sim 1182$ at 4 time points of adolescence) (Mascarell Maričić et al., 2020) by designing a robust and reliable ML pipeline. The goal of this study is to explore the relationship between adolescent brain and AAM using ML and discover any brain features that can be associated with AAM. As shown in Figure 1, we predict AAM at age 22 using brain morphometrics derived from structural imaging captured at three stages of adolescence – ages 14, 19, and 22. The structural features of different brain regions are extracted from two modalities of structural MRI, that is, T1-weighted imaging (T1w) and Diffusion Tensor Imaging (DTI). The most informative structural features for the ML model prediction are discovered using SHAP (Lundberg and Lee, 2017; Lundberg et al., 2020) to reveal the most distinct structural brain differences between AAMs and controls. Furthermore, we use multiple phenotypes of alcohol misuse such as the frequency of alcohol consumption, amount of consumption, onset of misuse, binge drinking, the AUDIT score, and other combinations, and systematically compare them. We also compare four different ML models, and multiple methods of controlling for confounds in ML and derive important methodological insights which are beneficial for reliably applying ML to psychiatric disorders such as AUD. To promote reproducibility and open science, the entire codebase used in this study, including the initial data analysis performed on the IMAGEN dataset are made available at https://github.com/RoshanRane/ML_for_IMAGEN(Rane and Kim, 2022; copy archived at swh:1:rev:6c493672ed700ded73c2b77e8976a5551921e634).

Figure 1

Download asset Open asset

The results are reported in the following four subsections: In subsection 1, different confound-control techniques are compared and the most suitable technique for this study is determined. Subsection 2 shows the results of the ML exploration performed with ten AAM labels, four ML models, and using imaging data from three time points of adolescence. This stage helps to determine the best phenotype of AAM and the best ML model. Subsection 3 reports the final results on the independent data ${}_{holdout}$ for all three time point analyses and subsection 4 shows the most informative features found in each of the analyses. Subsection 5 reports the result from the additional leave-one-site-out experiment.

Confound correction techniques

The sex $c_{sex}$ and recruitment site $c_{site}$ of subjects confound this study (refer to subsection 5.1 in ‘Materials and methods’) and their influence on the study needs to be controlled. We test three confound correction techniques on data ${}_{explore}$ – (a) confound regression (b) counterbalancing with undersampling and (c) counterbalancing with oversampling. To verify if these methods work as expected, the same analysis approach from Görgen et al., 2018 and the approach by Snoek et al., 2019 are employed. For the two confounds $c_{sex}$ and $c_{site}$, this requires us to test five input-output combinations ($X\to y$, $X\to c_{sex}$, $X\to c_{site}$, $c_{sex}\to y$ and $c_{site}\to y$) for a given $X\to y$ analysis.

Figure 2 shows the results of comparing different confound correction techniques for the ‘Binge’ phenotype. The following conclusions can be derived from this comparison:

Figure 2

Download asset Open asset

1. Sex and site can confound the AAM analysis: As shown in subplot (a), all the input-output combinations involving the confounds ($X\to c_{sex}$, $X\to c_{site}$, $c_{sex}\to y$ and $c_{site}\to y$) produce significant prediction accuracies before any confound correction is performed. This further adds to the evidence that both the confounds $c_{sex}$, $c_{site}$ can strongly influence the accuracy of the main analysis $X\to y$ and confound the analysis. 2. Confound regression is not a good choice when followed by a non-linear ML method: Following confound regression, the results of $X\to c_{sex}$ and $X\to c_{site}$ should become non-significant as the signal s_c has been removed from $X$. However, it is seen that in some cases the non-linear models SVM-rbf and GB are capable of detecting the confounding signal s_c from the imaging data. The red arrow in the subplot (b) points out one such case in the example shown. This is not surprising as the standard confound regression removes linear components of the signal s_c but does not remove any non-linear components that might still be present in $X$ (Görgen et al., 2018; Dinga et al., 2020). Furthermore, confound regression carries an additional risk of also regressing-out the useful signal in $X$ that does not confound the analysis $X\to y$ but is a co-variate of both $c$ and $y$ (Dinga et al., 2020). 3. Counterbalancing with oversampling is the best choice for this study: As expected, counterbalancing forces the $c_{sex}\to y$ and $c_{site}\to y$ accuracies to chance-level by removing the correlation between $c\sim y$ (subplots c and d). It can be seen that after the undersampled counterbalancing the results of the main analysis $X\to y$ also become non-significant as indicated by the red arrow in (c). This drastic reduction in performance is likely due to the reduction in the sample size of the training data by $n\sim 100-250$ from undersampling. Therefore, counterbalancing with oversampling of the minority group is a better alternative compared to undersampling.

This comparison was also repeated for two other AAM phenotypes - ‘Combined-seo’ and ‘Binge-growth’ and the above findings were found to be consistent across all of them. Hence, counterbalancing with oversampling is used as the confound-control technique in the main analysis. When performing over-sampled counterbalancing, it is ensured that the oversampling is done only for the training data.

ML exploration

The results from the ML exploration experiments are summarised in Figure 3. For the different AAM phenotypes, the balanced accuracies range between 45 and 73%. It must be noted that the results across different phenotypes are not directly comparable as each AAM phenotype classification task has a different sample size varying between $\approx 620-780$ (refer to ‘Materials and methods’ Table 2 and Appendix 1—table 2 for the list of phenotypes and their respective sample size). These differences in the number of samples in the two classes AAM and controls could add additional variance in the accuracy. Nevertheless, some useful observations can be made from the consistenties found across the three time point analyses, depicted in subplots (a), (b), and (c) of Figure 3:

Figure 3 with 1 supplement see all

Download asset Open asset

1. The most predictable phenotype from structural brain features for all three time point analyses is ‘Binge’ which measures the total lifetime experiences of being drunk from binge drinking.

2. Other individual phenotypes such as the amount of alcohol consumption (Amount), frequency of alcohol use (Frequency) and the age of AAM onset (Onset) are harder to predict from brain features compared to the binge drinking phenotype. The results on ‘Combined-seo’ and ‘Combined-ours’ shows that using phenotypes measuring amount and frequency of drinking in combination with binge drinking seems to also be detrimental to model performance.

3. All models perform poorly at predicting AAM phenotypes derived from AUDIT. This is surprising as AUDIT is considered a de facto screening test for measuring alcohol misuse (Kranzler and Soyka, 2018).

4. Among the four ML models, the SVM with non-linear kernel SVM-rbf, and the ensemble learning method GB perform better than the linear models LR and SVM-lin. This is further evident in the summary plot (d) in the figure.

Table 2

No.	Phenotype	Description	Questionnaire
1	Frequency	Number of occasions drinking alcohol in last 12 months	ESPAD 8b.
2	Amount	Number of alcohol drinks consumed on atypical drinking occasion	ESPAD prev31,AUDIT q2.
3	Onset	Had one or more binge-drinking experiences by the age of 14	ESPAD 29d
4	Binge	Total drunk episodes from binge-drinking in lifetime (by age 22)	ESPAD 19a,AUDIT q3.
5	Binge-growth	Longitudinal trajectory of binge-drinking experiences had per year	Growth curveof ESPAD 19b.
6	AUDIT	AUDIT screening test performed at the year of scan	AUDIT-total (q1-10).
7	AUDIT-quick	Only the first 3 questions of AUDIT screening test	AUDIT-freq (q1-3).
8	AUDIT-growth	Longitudinal changes in the AUDIT score measured over the years	Growth curve ofAUDIT-total.
9	Combined-seo	A combined risky-drinking phenotype from Seo et al., 2019 generated using amount, frequency, and binge-drinking data	ESPAD 8b, 17b, 19b,and TLFB alcohol2
10	Combined-ours	A combined risky-drinking phenotype developed by clusteringamount, frequency, and binge-drinking trajectory	AUDIT q1, q2,ESPAD 19a, growthcurve of ESPAD 19b.

In summary, the non-linear ML models SVM-rbf and GB coupled with the ‘Binge’ phenotype consistently perform the best in all three time point analyses. This is more clearly visible in the summary figure (d) where the results from all three analyses are combined in a single plot. Similar general observations can be made when the AUC-ROC metric is used to measure model performance (see Figure 3—figure supplement 1).

Generalization

The generalization test is performed with ‘Binge’ phenotype as the label and the two non-linear ML models, SVM-rbf and GB. The final results are shown in Figure 4. For the three analyses using imaging data from age 22, age 19, and age 14, as input, an average balanced accuracy of 78%, 75.5%, and 73.5% are achieved, respectively. Their average ROC-AUC scores are 83.93%, 83.1%, and 81.5% for the respective analyses. The accuracies for all three time point analyses are significant with ${\displaystyle p<0.01}$. To get a better intuition, please refer to Figure 4—figure supplement 1 that shows the model accuracies against the accuracies obtained from permutation tests.

Figure 4 with 1 supplement see all

Download asset Open asset

To further assess the causality in the $MR{I}_{age14}\to AA{M}_{age22}$ analysis, we repeated it by using only subjects who had no binge drinking experiences by age 14 ($n=477$) and also with subjects who had a maximum of one binge drinking experience ($n=565$) by age 14. The balanced accuracy obtained on the holdout set was $72.9\pm 2\%$ and $71.1\pm 2.3\%$, respectively.

Important brain regions

Following the generalization test, the most informative structural brain features are determined for the SVM-rbf model, as it performs relatively better among the two non-linear models tested on data ${}_{holdout}$ (see Figure 4). Figure 5 shows the list of the most important features for all three time point analyses and illustrates where they are located in the brain. It also shows whether these features have lower-than-average or higher-than-average values when the ML model predicts the subjects as AAMs.

Figure 5

Download asset Open asset

Several clusters of regions and feature values can be identified. Most of the important subcortical features are located around the lateral ventricles and the third ventricle and include CSF-related features such as the CSF mean-intensity, volume of left choroid plexus, and left corticospinal tract in the brain stem. Several white matter tracts are found to be informative such as parts of the corpus callosum, internal capsule, and posterior corona radiata. Furthermore, all of these white matter tracts, along with the brain stem have lower-than-average intensities in AAM predictions. The prominent cortical features are spread across the occipital, temporal, and frontal lobes. In the $MR{I}_{age22}\to AA{M}_{age22}$ analysis important cortical features appear in the occipital lobe. In contrast, for the future prediction analyses $MR{I}_{age19}\to AA{M}_{age22}$ and $MR{I}_{age14}\to AA{M}_{age22}$, clusters appear in the limbic system (parts of the cingulate cortex and right parahippocampal gyrus), frontal lobe (left-pars orbitalis, left-frontal pole, right-precentral gyrus, and left-rostral middle frontal gyrus) as well as in the temporal lobe (left-inferior temporal gyrus, left-temporal pole, and right-bank of the superior temporal sulcus). In the occipital lobe, AAMs predictions have lower grey matter thickness in the right-cuneus, lateral occipital, and pericalcarine cortices, and higher curvature index in left-cuneus and left-pericalcarine cortex. The list of all the informative features are provided in Appendix 1—table 3 along with their feature type, modality, and respective SHAP values in each CV folds.

Cross-site experiment

The result from the leave-one-site-out CV experiment are shown in Figure 6. The ML models perform close-to-chance for all AAM labels in the ML exploration experiments and fail to produce a significant performance for any of the three time points in the generalization test. For the ‘Binge’ label in the ML exploration stage, the model accuracy displays very high variance, as compared to the main experiment (compare Figure 6 with Figure 3 (d)). This suggests that the performance of the ML models varies greatly across sites in this study.

Figure 6

Download asset Open asset

For over two decades, researchers have tried to uncover the relationship that exist between adolescent alcohol misuse (AAM) and brain development. Many previous studies found that such a relationship exists (see Table 1) but with low-to-medium effect size (Nagel et al., 2005; Whelan et al., 2014; Squeglia et al., 2017; Seo et al., 2019; De Bellis et al., 2005; McQueeny et al., 2009; Luciana et al., 2013). The brain regions linked with AAM varied greatly across studies (see highlighted text in Table 1). This inconsistency in findings and effect sizes could be due to methodological limitations, small sample studies, unavailability of long-term longitudinal data like IMAGEN (Mascarell Maričić et al., 2020), or simply due to the heterogeneous expression of AAM in the brain. In our study, ML models predicted AAM with significantly above-chance accuracies in the range $73.1\%-78\%$ (ROC-AUC in $81.5\%-83.9\%$) from adolescent brain structure captured at ages 14, 19, and 22. Thus, our results demonstrate that adolescent brain structure is indeed associated with alcohol misuse during this period.

The causality of the relationship between adolescent brain structure and AAM is not clear (Whelan et al., 2014; Robert et al., 2020). The relationship could arise from alcohol misuse inducing neurotoxicity (Zahr and Pfefferbaum, 2017) causing the observed changes in their brains. It could also be that these structural differences precede AAM and such adolescents are just more vulnerable towards alcohol misuse (Chambers et al., 2003; Sanchez-Roige et al., 2019). Such neuropsychological predisposition could stem from genetic predispositions or from influencing environmental factors such as early stress or childhood trauma (Baker et al., 2013; Ross et al., 2021), misuse of other drugs such as cannabis (French et al., 2015) and tobacco, and parental drug misuse (Jones and Nagel, 2019). There might also be an interaction effect between alcohol-induced neurotoxity and environmental and genetic predispositions (Robert et al., 2020). While the direction of causality is still under active investigation (Robert et al., 2020; Bourque et al., 2016), the significantly high accuracies obtained in our study for $MR{I}_{age19}\to AA{M}_{age22}$ and especially $MR{I}_{age14}\to AA{M}_{age22}$ suggest that these structural differences might be preceding alcohol misuse behavior. Out of the 265 subjects that took the ESPAD survey at age 14 and belonged to the AAM category in $MR{I}_{age14}\to AA{M}_{age22}$ analysis, 83.3% of subjects reported having no or just one binge drinking experience until age 14. When we repeated the $MR{I}_{age14}\to AA{M}_{age22}$ analysis with only the subjects who had no binge drinking experiences ($n=477$) or a maximum of one binge drinking experience ($n=565$) by age 14, we obtained a balanced accuracy of $72.9\pm 2\%$ and $71.1\pm 2.3\%$ respectively, on the holdout data. This is comparable to the main result of $73.1\pm 2\%$. This result provides further evidence for the findings of Robert et al., 2020 that certain cerebral predispositions might precede alcohol abuse in adolescents. Thus, like (Robert et al., 2020) we also advocate caution when interpreting the results from previous cross-sectional studies suggesting alcohol-induced brain atrophy. We identified the most informative brain features for the ML predictions using SHAP that has been successfully applied to medical data (Lundberg and Lee, 2017; Lundberg et al., 2020; Molnar, 2022). The important features were found to be distributed across several subcortical and cortical regions of the brain, implying that the association between AAM and brain structure is widespread and heterogeneous. In accordance with previous studies, AAM was associated with lower DTI-FA intensities in several white matter tracts and the brain stem (McQueeny et al., 2009; Jacobus et al., 2013; Jones et al., 2018) and reduced GM thickness (Squeglia et al., 2017; Pfefferbaum et al., 2018), especially in the occipital lobe. Features of anterior cingulate cortex (Squeglia et al., 2017; Seo et al., 2019; Jones et al., 2018), middle frontal and precentral gyrus (Luciana et al., 2013), hippocampus (De Bellis et al., 2000; Nagel et al., 2005), and right parahippocampal gyrus (Whelan et al., 2014) were also found to be informative, although the type of feature and the average feature value in AAMs differed from previous studies. Features from the frontal lobe and cerebellum were informative only for future AAM (Jones and Nagel, 2019) but not for current AAM prediction, in contrast to findings of De Bellis et al., 2005; Whelan et al., 2014; Seo et al., 2019. This difference could be due to the meticulous confound control performed in this study for sex and site of the subjects. Additionally, our ML models also found CSF-related features in the third and lateral ventricles, and some regions of the temporal cortex as informative features for AAM prediction.

In the ML exploration stage, we found that the binge drinking phenotype, which is commonly used in previous studies (Nagel et al., 2005; Whelan et al., 2014; Robert et al., 2020), was the most predictable phenotype of AAM as compared to frequency, amount, or onset of alcohol misuse. Curiously, phenotypes derived from AUDIT, which is a gold standard of screening for alcohol misuse (Kranzler and Soyka, 2018), did not score significantly above-chance in any of the three time point analyses. Other similar compound metrics that use measures of alcohol use frequency and amount along with binge drinking, such as ‘Combined-seo’ and ‘Combined-ours’, also perform worse than using just the binge drinking information. This suggests that using other phenotypes of alcohol misuse in combination with binge drinking was detrimental to the prediction task, as compared to using only binge drinking. Different phenotypes of AAM capture slightly different psychosocial characteristics of adolescents (Castellanos-Ryan et al., 2013). For instance, ‘Amount’ correlates significantly with agreeableness and a life history of relocation valence ($r=-0.14$, ${\displaystyle p<0.001}$), accident valence ($r=-0.16$, ${\displaystyle p<0.001}$) and sexuality frequency ($r=-0.17$, ${\displaystyle p<0.001}$), whereas the other phenotypes do not (${\displaystyle p>0.01}$). ‘AUDIT’ and it’s derivatives significantly correlate with impulsivity trait ($r=0.23$, ${\displaystyle p<0.001}$) on SURPS, where as ‘Binge’ does not ($r=0.09$, ${\displaystyle p>0.01}$) but they both correlate with sensation seeking trait (${\displaystyle r>0.29}$, ${\displaystyle p<0.001}$) as also found in previous studies (Castellanos-Ryan et al., 2011). Castellanos-Ryan et al., 2013 have found that these two traits manifest differently in the brain. Therefore, one can hypothesize that the psychosocial differences and their associated neural correlates (Castellanos-Ryan et al., 2011) between ‘Binge’ and the other AAM phenotypes might explain the $2-10\%$ higher accuracy obtained with ‘Binge’.

In contrast to the main results, the ML models failed to attain significantly high prediction accuracy in the leave-one-site-out experiment as the scores displayed high variance across the CV folds (refer to Figure 6). On further investigation, we found that the ML models performed especially poorly on test data from Dublin and Nottingham ($\le 60\%$ balanced accuracy) across all time points and metrics. On the contrary, models always performed better-than-chance on subjects from Dresden, Mannheim, and Hamburg. When we compared this with the main experiment, a similar pattern was found. The models least generalized to test subjects from the sites Dublin and Nottingham, across all 7 CV folds. Notably, the accuracy across sites did not correlate with the sample size of the sites, the ratio of AAMs to controls in the site, or their sex distribution. The results are shown in Appendix 1—figure 2 and Appendix 1—figure 3. Altogether, these results suggest that the relationship discovered in this study performs diversely on subjects from different sites and does not generalize equally across all sites of the IMAGEN dataset.

Methodological insights: To the best of our knowledge, this is the first study to analyze and reports results on the complete longitudinal data from IMAGEN, including the follow-up 3 data. Two previous studies, (Whelan et al., 2014; Seo et al., 2019) performed similar ML analysis on the IMAGEN data and unlike us, found only a weak association between structural imaging and AAM. The logistic regression model in Whelan et al., 2014 scored $58\pm 8\%$ ROC-AUC when predicting AAM at age 14 from structural imaging features collected at age 14 (BL) and $63\pm 7\%$ ROC-AUC at predicting AAM at age 16 (FU1). This lower accuracy with high variance obtained in their experiments can be attributed to - (a) the relatively smaller sample size used in their study ($n\sim 265-271$), (b) unavailability of long-term AAM information from IMAGEN’s FU2 and FU3 data, (c) using only a linear ML model, and (d) only using GM volume and thickness as structural features. On the other hand, Seo et al., 2019’s models achieved accuracies in the range $56-58\%$ when predicting AAM at age 19 (FU2) using imaging features from age 19, and did not get a significant accuracy when they used imaging features from age 14. This lower performance can be attributed to the following experimental design decisions - (a) Seo et al., 2019 used GM volume and thickness features from just 24 regions of the brain associated to cue-reactivity, (b) their AAM phenotype is not the best phenotype of AAM as evident from the results of our ML exploration (see results for ‘Combined-seo’ in Figure 3), and (c) the confound-control technique used in their study, confound regression, can result in under-performance as demonstrated in Figure 2.

In contrast to these previous works, our study has the following advantages: First, we use 719 structural features extracted from 2 MRI modalities, T1w and DTI, that include not only GM volume and thickness but also surface area, curvature, and WM and GM intensities from all cortical and sub-cortical regions in the brains. Second, we empirically derive the best AAM label for the task by comparing different phenotypes previously used in the literature. For the different AAM phenotypes, the balanced accuracies range between chance to significant performance ($45\%-73\%$), emphasizing the importance of the choice of the label in such ML studies with low effect sizes. And finally, we test different confound correction techniques and use the one that effectively controls for the influence of confounds without also destroying the signal of interest. In summary, the higher accuracy in the current study can be attributed to not just the availability of long-term data on AAM but also to the rigorous comparison of different labels of AAM, different ML models and confound control techniques.

Among the four different ML models tested, the two non-linear models, SVM-rbf and GB, consistently performed better than the two linear models. We also explicitly ensured that the confounding influence of sex and site were eliminated by combining suggestions from Görgen et al., 2018 and Snoek et al., 2019. We found evidence that the linear confound regression technique used often in previous ML-based neuroimaging studies (Seo et al., 2019; Robert et al., 2020; Snoek et al., 2019), might not be the best choice as it cannot be used with non-linear models such as SVM-rbf or Naive Bayes used in Seo et al., 2019 and distorts the signal of interest from the neuroimaging data (Dinga et al., 2020) as seen in Figure 2. In contrast, counterbalancing using oversampling is recommended as it successfully removed the influence of the confounds without reducing the sample size in the study.

Future work: An important follow-up work would involve further investigating the association we found between AAM and adolescent brain structure and its clinical implications. For instance, one can analyze if certain environmental risk factors such as childhood abuse, parental drug use, or life event stressors mediate the relationship we found between brain structure and AAM behavior in the IMAGEN cohort. Another direction would be to further investigate the brain features associated with AAM and understand the relative contributions of specific brain networks (for example, similar to Seo et al., 2019) and certain specific feature types such as thickness, or volumes. Specifically, since ML feature attribution methods such as SHAP can be misled by the presence of correlated features (Molnar, 2022; Lundberg and Lee, 2017), it would be necessary to before-hand determine which features might be correlated and either exclude them, or permute correlated features together in groups when computing SHAP values (Molnar, 2022). Another important future work would be to reproduce our findings on another data set such as NCANDA (Brown et al., 2015) comprising adolescent subjects from a different geographic region. It would also be interesting to explore other modalities such as functional connectivity (fMRI) to predict AAM (Ruan et al., 2019).

This is a computational study. All data analyses code including the modelling pipeline are openly provided publicly at https://github.com/RoshanRane/ML_for_IMAGEN, (copy archived at swh:1:rev:6c493672ed700ded73c2b77e8976a5551921e634) for reuse and reproduction. Approval to use the IMAGEN dataset for this study was provided under the approval username / project code 'edeman'.

1. 1. Association WM

   (2013) World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects

   JAMA 310:2191–2194.

   https://doi.org/10.1001/jama.2013.281053

   • PubMed
   • Google Scholar
2. 1. Baker LM
   2. Williams LM
   3. Korgaonkar MS
   4. Cohen RA
   5. Heaps JM
   6. Paul RH

   (2013) Impact of early vs. late childhood early life stress on brain morphometrics

   Brain Imaging and Behavior 7:196–203.

   https://doi.org/10.1007/s11682-012-9215-y

   • PubMed
   • Google Scholar
3. 1. Bengio Y
   2. Grandvalet Y

   (2004)

   No unbiased estimator of the variance of k-fold cross-validation

   Journal of Machine Learning Research 5:1089–1105.

   • Google Scholar
4. Conference

   1. Boser BE
   2. Guyon IM
   3. Vapnik VN

   (1992)

   A training algorithm for optimal margin classifiers

   In Proceedings of the fifth annual workshop on Computational learning theory. pp. 144–152.

   • Google Scholar
5. 1. Bourque J
   2. Baker TE
   3. Dagher A
   4. Evans AC
   5. Garavan H
   6. Leyton M
   7. Séguin JR
   8. Pihl R
   9. Conrod PJ

   (2016) Effects of delaying binge drinking on adolescent brain development: a longitudinal neuroimaging study

   BMC Psychiatry 16:1–9.

   https://doi.org/10.1186/s12888-016-1148-3

   • PubMed
   • Google Scholar
6. 1. Brown SA
   2. Brumback T
   3. Tomlinson K
   4. Cummins K
   5. Thompson WK
   6. Nagel BJ
   7. De Bellis MD
   8. Hooper SR
   9. Clark DB
   10. Chung T
   11. Hasler BP
   12. Colrain IM
   13. Baker FC
   14. Prouty D
   15. Pfefferbaum A
   16. Sullivan EV
   17. Pohl KM
   18. Rohlfing T
   19. Nichols BN
   20. Chu W
   21. Tapert SF

   (2015) The National Consortium on Alcohol and NeuroDevelopment in Adolescence (NCANDA): A Multisite Study of Adolescent Development and Substance Use

   Journal of Studies on Alcohol and Drugs 76:895–908.

   https://doi.org/10.15288/jsad.2015.76.895

   • PubMed
   • Google Scholar
7. 1. Button KS
   2. Ioannidis JP
   3. Mokrysz C
   4. Nosek BA
   5. Flint J
   6. Robinson ES
   7. Munafò MR

   (2013) Power failure: why small sample size undermines the reliability of neuroscience

   Nature Reviews. Neuroscience 14:365–376.

   https://doi.org/10.1038/nrn3475

   • PubMed
   • Google Scholar
8. 1. Castellanos-Ryan N
   2. Rubia K
   3. Conrod PJ

   (2011) Response inhibition and reward response bias mediate the predictive relationships between impulsivity and sensation seeking and common and unique variance in conduct disorder and substance misuse

   Alcoholism, Clinical and Experimental Research 35:140–155.

   https://doi.org/10.1111/j.1530-0277.2010.01331.x

   • PubMed
   • Google Scholar
9. 1. Castellanos-Ryan N
   2. O’Leary-Barrett M
   3. Sully L
   4. Conrod P

   (2013) Sensitivity and specificity of a brief personality screening instrument in predicting future substance use, emotional, and behavioral problems: 18-month predictive validity of the Substance Use Risk Profile Scale

   Alcoholism, Clinical and Experimental Research 37 Suppl 1:E281–E290.

   https://doi.org/10.1111/j.1530-0277.2012.01931.x

   • PubMed
   • Google Scholar
10. 1. Chambers RA
    2. Taylor JR
    3. Potenza MN

    (2003) Developmental neurocircuitry of motivation in adolescence: a critical period of addiction vulnerability

    The American Journal of Psychiatry 160:1041–1052.

    https://doi.org/10.1176/appi.ajp.160.6.1041

    • PubMed
    • Google Scholar
11. 1. Chapelle O
    2. Vapnik V
    3. Bousquet O
    4. Mukherjee S

    (2002) Choosing multiple parameters for support vector machines

    Machine Learning 46:131–159.

    https://doi.org/10.1023/A:1012450327387

    • Google Scholar
12. Conference

    1. Chen T
    2. Guestrin C

    (2016)

    Xgboost: A scalable tree boosting system

    In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining. pp. 785–794.

    • Google Scholar
13. 1. Crews F
    2. He J
    3. Hodge C

    (2007) Adolescent cortical development: a critical period of vulnerability for addiction

    Pharmacology, Biochemistry, and Behavior 86:189–199.

    https://doi.org/10.1016/j.pbb.2006.12.001

    • PubMed
    • Google Scholar
14. 1. De Bellis MD
    2. Clark DB
    3. Beers SR
    4. Soloff PH
    5. Boring AM
    6. Hall J
    7. Kersh A
    8. Keshavan MS

    (2000) Hippocampal volume in adolescent-onset alcohol use disorders

    The American Journal of Psychiatry 157:737–744.

    https://doi.org/10.1176/appi.ajp.157.5.737

    • PubMed
    • Google Scholar
15. 1. De Bellis MD
    2. Narasimhan A
    3. Thatcher DL
    4. Keshavan MS
    5. Soloff P
    6. Clark DB

    (2005) Prefrontal cortex, thalamus, and cerebellar volumes in adolescents and young adults with adolescent-onset alcohol use disorders and comorbid mental disorders

    Alcoholism, Clinical and Experimental Research 29:1590–1600.

    https://doi.org/10.1097/01.alc.0000179368.87886.76

    • PubMed
    • Google Scholar
16. 1. Deeken F
    2. Banaschewski T
    3. Kluge U
    4. Rapp MA

    (2020) Risk and Protective Factors for Alcohol Use Disorders Across the Lifespan

    Current Addiction Reports 7:245–251.

    https://doi.org/10.1007/s40429-020-00313-z

    • Google Scholar
17. 1. DeWit DJ
    2. Adlaf EM
    3. Offord DR
    4. Ogborne AC

    (2000) Age at first alcohol use: a risk factor for the development of alcohol disorders

    The American Journal of Psychiatry 157:745–750.

    https://doi.org/10.1176/appi.ajp.157.5.745

    • PubMed
    • Google Scholar
18. Preprint

    1. Dinga R
    2. Schmaal L
    3. Penninx BW
    4. Veltman DJ
    5. Marquand AF

    (2020) Controlling for Effects of Confounding Variables on Machine Learning Predictions

    bioRxiv.

    https://doi.org/10.1101/2020.08.17.255034

    • Google Scholar
19. 1. Fan R-E
    2. Chang K-W
    3. Hsieh C-J
    4. Wang X-R
    5. Lin C-J

    (2008)

    Liblinear: A library for large linear classification

    The Journal of Machine Learning Research 9:1871–1874.

    • Google Scholar
20. 1. Filippi I
    2. Galinowski A
    3. Lemaître H
    4. Massot C
    5. Zille P
    6. Frère P
    7. Miranda-Marcos R
    8. Trichard C
    9. Guldner S
    10. Vulser H
    11. Paillère-Martinot M-L
    12. Quinlan EB
    13. Desrivieres S
    14. Gowland P
    15. Bokde A
    16. Garavan H
    17. Heinz A
    18. Walter H
    19. Daedelow L
    20. Büchel C
    21. Bromberg U
    22. Conrod PJ
    23. Flor H
    24. Banaschewski T
    25. Nees F
    26. Heintz S
    27. Smolka M
    28. Vetter NC
    29. Papadopoulos-Orfanos D
    30. Whelan R
    31. Poustka L
    32. Paus T
    33. Schumann G
    34. Artiges E
    35. Martinot J-L
    36. IMAGEN Consortium

    (2021) Neuroimaging evidence for structural correlates in adolescents resilient to polysubstance use: A five-year follow-up study

    European Neuropsychopharmacology 49:11–22.

    https://doi.org/10.1016/j.euroneuro.2021.03.001

    • PubMed
    • Google Scholar
21. 1. French L
    2. Gray C
    3. Leonard G
    4. Perron M
    5. Pike GB
    6. Richer L
    7. Séguin JR
    8. Veillette S
    9. Evans CJ
    10. Artiges E
    11. Banaschewski T
    12. Bokde AWL
    13. Bromberg U
    14. Bruehl R
    15. Buchel C
    16. Cattrell A
    17. Conrod PJ
    18. Flor H
    19. Frouin V
    20. Gallinat J
    21. Garavan H
    22. Gowland P
    23. Heinz A
    24. Lemaitre H
    25. Martinot J-L
    26. Nees F
    27. Orfanos DP
    28. Pangelinan MM
    29. Poustka L
    30. Rietschel M
    31. Smolka MN
    32. Walter H
    33. Whelan R
    34. Timpson NJ
    35. Schumann G
    36. Smith GD
    37. Pausova Z
    38. Paus T

    (2015) Early Cannabis Use, Polygenic Risk Score for Schizophrenia and Brain Maturation in Adolescence

    JAMA Psychiatry 72:1002–1011.

    https://doi.org/10.1001/jamapsychiatry.2015.1131

    • PubMed
    • Google Scholar
22. 1. Friedman JH

    (2001) Greedy function approximation: A gradient boosting machine

    The Annals of Statistics 29:03451.

    https://doi.org/10.1214/aos/1013203451

    • Google Scholar
23. 1. Giedd JN

    (2004) Structural magnetic resonance imaging of the adolescent brain

    Annals of the New York Academy of Sciences 1021:77–85.

    https://doi.org/10.1196/annals.1308.009

    • PubMed
    • Google Scholar
24. 1. Görgen K
    2. Hebart MN
    3. Allefeld C
    4. Haynes J-D

    (2018) The same analysis approach: Practical protection against the pitfalls of novel neuroimaging analysis methods

    NeuroImage 180:19–30.

    https://doi.org/10.1016/j.neuroimage.2017.12.083

    • PubMed
    • Google Scholar
25. 1. Grant JD
    2. Scherrer JF
    3. Lynskey MT
    4. Lyons MJ
    5. Eisen SA
    6. Tsuang MT
    7. True WR
    8. Bucholz KK

    (2006) Adolescent alcohol use is a risk factor for adult alcohol and drug dependence: evidence from a twin design

    Psychological Medicine 36:109–118.

    https://doi.org/10.1017/S0033291705006045

    • PubMed
    • Google Scholar
26. 1. Grant BF
    2. Goldstein RB
    3. Saha TD
    4. Chou SP
    5. Jung J
    6. Zhang H
    7. Pickering RP
    8. Ruan WJ
    9. Smith SM
    10. Huang B
    11. Hasin DS

    (2015) Epidemiology of DSM-5 Alcohol Use Disorder: Results From the National Epidemiologic Survey on Alcohol and Related Conditions III

    JAMA Psychiatry 72:757–766.

    https://doi.org/10.1001/jamapsychiatry.2015.0584

    • PubMed
    • Google Scholar
27. 1. Hebart MN
    2. Baker CI

    (2018) Deconstructing multivariate decoding for the study of brain function

    NeuroImage 180:4–18.

    https://doi.org/10.1016/j.neuroimage.2017.08.005

    • PubMed
    • Google Scholar
28. 1. Ioannidis JP

    (2005) Why most published research findings are false

    PLOS Medicine 2:e124.

    https://doi.org/10.1371/journal.pmed.0020124

    • PubMed
    • Google Scholar
29. 1. Jacobus J
    2. Squeglia LM
    3. Bava S
    4. Tapert SF

    (2013) White matter characterization of adolescent binge drinking with and without co-occurring marijuana use: a 3-year investigation

    Psychiatry Research 214:374–381.

    https://doi.org/10.1016/j.pscychresns.2013.07.014

    • PubMed
    • Google Scholar
30. 1. Jia T
    2. Xie C
    3. Banaschewski T
    4. Barker GJ
    5. Bokde ALW
    6. Büchel C
    7. Quinlan EB
    8. Desrivières S
    9. Flor H
    10. Grigis A
    11. Garavan H
    12. Gowland P
    13. Heinz A
    14. Ittermann B
    15. Martinot J-L
    16. Martinot M-LP
    17. Nees F
    18. Orfanos DP
    19. Poustka L
    20. Fröhner JH
    21. Smolka MN
    22. Walter H
    23. Whelan R
    24. Schumann G
    25. Robbins TW
    26. Feng J
    27. IMAGEN Consortium

    (2021) Neural network involving medial orbitofrontal cortex and dorsal periaqueductal gray regulation in human alcohol abuse

    Science Advances 7:eabd4074.

    https://doi.org/10.1126/sciadv.abd4074

    • PubMed
    • Google Scholar
31. 1. Jones DK
    2. Williams SCR
    3. Gasston D
    4. Horsfield MA
    5. Simmons A
    6. Howard R

    (2002) Isotropic resolution diffusion tensor imaging with whole brain acquisition in a clinically acceptable time

    Human Brain Mapping 15:216–230.

    https://doi.org/10.1002/hbm.10018

    • PubMed
    • Google Scholar
32. 1. Jones SA
    2. Lueras JM
    3. Nagel BJ

    (2018)

    Effects of Binge Drinking on the Developing Brain

    Alcohol Research 39:87–96.

    • PubMed
    • Google Scholar
33. 1. Jones SA
    2. Nagel BJ

    (2019) Altered frontostriatal white matter microstructure is associated with familial alcoholism and future binge drinking in adolescence

    Neuropsychopharmacology : Official Publication of the American College of Neuropsychopharmacology 44:1076–1083.

    https://doi.org/10.1038/s41386-019-0315-x

    • PubMed
    • Google Scholar
34. 1. Kranzler HR
    2. Soyka M

    (2018) Diagnosis and Pharmacotherapy of Alcohol Use Disorder: A Review

    JAMA 320:815–824.

    https://doi.org/10.1001/jama.2018.11406

    • PubMed
    • Google Scholar
35. 1. Kriegeskorte N
    2. Simmons WK
    3. Bellgowan PSF
    4. Baker CI

    (2009) Circular analysis in systems neuroscience: the dangers of double dipping

    Nature Neuroscience 12:535–540.

    https://doi.org/10.1038/nn.2303

    • PubMed
    • Google Scholar
36. 1. Kühn S
    2. Mascharek A
    3. Banaschewski T
    4. Bodke A
    5. Bromberg U
    6. Büchel C
    7. Quinlan EB
    8. Desrivieres S
    9. Flor H
    10. Grigis A
    11. Garavan H
    12. Gowland PA
    13. Heinz A
    14. Ittermann B
    15. Martinot J-L
    16. Nees F
    17. Papadopoulos Orfanos D
    18. Paus T
    19. Poustka L
    20. Millenet S
    21. Fröhner JH
    22. Smolka MN
    23. Walter H
    24. Whelan R
    25. Schumann G
    26. Lindenberger U
    27. Gallinat J
    28. IMAGEN Consortium

    (2019) Predicting development of adolescent drinking behaviour from whole brain structure at 14 years of age

    eLife 8:e44056.

    https://doi.org/10.7554/eLife.44056

    • PubMed
    • Google Scholar
37. 1. Lebel C
    2. Beaulieu C

    (2011) Longitudinal development of human brain wiring continues from childhood into adulthood

    The Journal of Neuroscience 31:10937–10947.

    https://doi.org/10.1523/JNEUROSCI.5302-10.2011

    • PubMed
    • Google Scholar
38. 1. Lindquist MA
    2. Mejia A

    (2015) Zen and the art of multiple comparisons

    Psychosomatic Medicine 77:114–125.

    https://doi.org/10.1097/PSY.0000000000000148

    • PubMed
    • Google Scholar
39. 1. Luciana M
    2. Collins PF
    3. Muetzel RL
    4. Lim KO

    (2013) Effects of alcohol use initiation on brain structure in typically developing adolescents

    The American Journal of Drug and Alcohol Abuse 39:345–355.

    https://doi.org/10.3109/00952990.2013.837057

    • PubMed
    • Google Scholar
40. Conference

    1. Lundberg SM
    2. Lee S-I

    (2017)

    A unified approach to interpreting model predictions

    Proceedings of the 31st international conference on neural information processing systems. pp. 4768–4777.

    • Google Scholar
41. 1. Lundberg SM
    2. Erion G
    3. Chen H
    4. DeGrave A
    5. Prutkin JM
    6. Nair B
    7. Katz R
    8. Himmelfarb J
    9. Bansal N
    10. Lee S-I

    (2020) From Local Explanations to Global Understanding with Explainable AI for Trees

    Nature Machine Intelligence 2:56–67.

    https://doi.org/10.1038/s42256-019-0138-9

    • PubMed
    • Google Scholar
42. 1. Mascarell Maričić L
    2. Walter H
    3. Rosenthal A
    4. Ripke S
    5. Quinlan EB
    6. Banaschewski T
    7. Barker GJ
    8. Bokde ALW
    9. Bromberg U
    10. Büchel C
    11. Desrivières S
    12. Flor H
    13. Frouin V
    14. Garavan H
    15. Itterman B
    16. Martinot J-L
    17. Martinot M-LP
    18. Nees F
    19. Orfanos DP
    20. Paus T
    21. Poustka L
    22. Hohmann S
    23. Smolka MN
    24. Fröhner JH
    25. Whelan R
    26. Kaminski J
    27. Schumann G
    28. Heinz A
    29. IMAGEN consortium

    (2020) The IMAGEN study: A decade of imaging genetics in adolescents

    Molecular Psychiatry 25:2648–2671.

    https://doi.org/10.1038/s41380-020-0822-5

    • PubMed
    • Google Scholar
43. 1. McQueeny T
    2. Schweinsburg BC
    3. Schweinsburg AD
    4. Jacobus J
    5. Bava S
    6. Frank LR
    7. Tapert SF

    (2009) Altered white matter integrity in adolescent binge drinkers

    Alcoholism, Clinical and Experimental Research 33:1278–1285.

    https://doi.org/10.1111/j.1530-0277.2009.00953.x

    • PubMed
    • Google Scholar
44. Book

    1. Molnar C

    (2022)

    Interpretable Machine Learning: A Guide for Making Black Box Models Explainable (Second Edition)

    self-published.

    • Google Scholar
45. 1. Monti PM
    2. Miranda R
    3. Nixon K
    4. Sher KJ
    5. Swartzwelder HS
    6. Tapert SF
    7. White A
    8. Crews FT

    (2005) Adolescence: booze, brains, and behavior

    Alcoholism, Clinical and Experimental Research 29:207–220.

    https://doi.org/10.1097/01.alc.0000153551.11000.f3

    • PubMed
    • Google Scholar
46. 1. Nagel BJ
    2. Schweinsburg AD
    3. Phan V
    4. Tapert SF

    (2005) Reduced hippocampal volume among adolescents with alcohol use disorders without psychiatric comorbidity

    Psychiatry Research 139:181–190.

    https://doi.org/10.1016/j.pscychresns.2005.05.008

    • PubMed
    • Google Scholar
47. 1. Nixon K
    2. McClain JA

    (2010) Adolescence as a critical window for developing an alcohol use disorder: current findings in neuroscience

    Current Opinion in Psychiatry 23:227–232.

    https://doi.org/10.1097/YCO.0b013e32833864fe

    • PubMed
    • Google Scholar
48. 1. Ojala M
    2. Garriga GC

    (2010)

    Permutation tests for studying classifier performance

    Journal of Machine Learning Research 11:1833–1863.

    • Google Scholar
49. 1. Pfefferbaum A
    2. Kwon D
    3. Brumback T
    4. Thompson WK
    5. Cummins K
    6. Tapert SF
    7. Brown SA
    8. Colrain IM
    9. Baker FC
    10. Prouty D
    11. De Bellis MD
    12. Clark DB
    13. Nagel BJ
    14. Chu W
    15. Park SH
    16. Pohl KM
    17. Sullivan EV

    (2018) Altered Brain Developmental Trajectories in Adolescents After Initiating Drinking

    The American Journal of Psychiatry 175:370–380.

    https://doi.org/10.1176/appi.ajp.2017.17040469

    • PubMed
    • Google Scholar
50. Software

    1. Rane RP
    2. Kim JH

    (2022) ML_for_IMAGEN, version swh:1:rev:6c493672ed700ded73c2b77e8976a5551921e634

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:087b3e0b49221fbf1e8e145e0b79ba5a856ab457;origin=https://github.com/RoshanRane/ML_for_IMAGEN;visit=swh:1:snp:f48b26d4ce0ce39ba38965697100f63132274db0;anchor=swh:1:rev:6c493672ed700ded73c2b77e8976a5551921e634
51. 1. Rao A
    2. Monteiro JM
    3. Mourao-Miranda J

    (2017) Alzheimer’s Disease Initiative, et al Predictive modelling using neuroimaging data in the presence of confounds

    NeuroImage 150:23–49.

    https://doi.org/10.1016/j.neuroimage.2017.01.066

    • Google Scholar
52. 1. Robert GH
    2. Luo Q
    3. Yu T
    4. Chu C
    5. Ing A
    6. Jia T
    7. Papadopoulos Orfanos D
    8. Burke-Quinlan E
    9. Desrivières S
    10. Ruggeri B
    11. Spechler P
    12. Chaarani B
    13. Tay N
    14. Banaschewski T
    15. Bokde ALW
    16. Bromberg U
    17. Flor H
    18. Frouin V
    19. Gowland P
    20. Heinz A
    21. Ittermann B
    22. Martinot J-L
    23. Paillère Martinot M-L
    24. Nees F
    25. Poustka L
    26. Smolka MN
    27. Vetter NC
    28. Walter H
    29. Whelan R
    30. Conrod P
    31. Barker T
    32. Garavan H
    33. Schumann G
    34. IMAGEN Consortium

    (2020) Association of Gray Matter and Personality Development With Increased Drunkenness Frequency During Adolescence

    JAMA Psychiatry 77:409–419.

    https://doi.org/10.1001/jamapsychiatry.2019.4063

    • PubMed
    • Google Scholar
53. 1. Ross MC
    2. Dvorak D
    3. Sartin-Tarm A
    4. Botsford C
    5. Cogswell I
    6. Hoffstetter A
    7. Putnam O
    8. Schomaker C
    9. Smith P
    10. Stalsberg A
    11. Wang Y
    12. Xiong M
    13. Cisler JM

    (2021) Gray matter volume correlates of adolescent posttraumatic stress disorder: A comparison of manual intervention and automated segmentation in FreeSurfer

    Psychiatry Research. Neuroimaging 313:111297.

    https://doi.org/10.1016/j.pscychresns.2021.111297

    • PubMed
    • Google Scholar
54. 1. Rozycki M
    2. Satterthwaite TD
    3. Koutsouleris N
    4. Erus G
    5. Doshi J
    6. Wolf DH
    7. Fan Y
    8. Gur RE
    9. Gur RC
    10. Meisenzahl EM
    11. Zhuo C
    12. Yin H
    13. Yan H
    14. Yue W
    15. Zhang D
    16. Davatzikos C

    (2018) Multisite Machine Learning Analysis Provides a Robust Structural Imaging Signature of Schizophrenia Detectable Across Diverse Patient Populations and Within Individuals

    Schizophrenia Bulletin 44:1035–1044.

    https://doi.org/10.1093/schbul/sbx137

    • PubMed
    • Google Scholar
55. 1. Ruan H
    2. Zhou Y
    3. Luo Q
    4. Robert GH
    5. Desrivières S
    6. Quinlan EB
    7. Liu Z
    8. Banaschewski T
    9. Bokde ALW
    10. Bromberg U
    11. Büchel C
    12. Flor H
    13. Frouin V
    14. Garavan H
    15. Gowland P
    16. Heinz A
    17. Ittermann B
    18. Martinot J-L
    19. Martinot M-LP
    20. Nees F
    21. Orfanos DP
    22. Poustka L
    23. Hohmann S
    24. Fröhner JH
    25. Smolka MN
    26. Walter H
    27. Whelan R
    28. Li F
    29. Schumann G
    30. Feng J
    31. IMAGEN Consortium

    (2019) Adolescent binge drinking disrupts normal trajectories of brain functional organization and personality maturation

    NeuroImage. Clinical 22:101804.

    https://doi.org/10.1016/j.nicl.2019.101804

    • PubMed
    • Google Scholar
56. 1. Sanchez-Roige S
    2. Palmer AA
    3. Fontanillas P
    4. Elson SL
    5. Adams MJ
    6. Howard DM
    7. Edenberg HJ
    8. Davies G
    9. Crist RC
    10. Deary IJ
    11. McIntosh AM
    12. Clarke T-K
    13. 23andMe Research Team, the Substance Use Disorder Working Group of the Psychiatric Genomics Consortium

    (2019) Genome-Wide Association Study Meta-Analysis of the Alcohol Use Disorders Identification Test (AUDIT) in Two Population-Based Cohorts

    The American Journal of Psychiatry 176:107–118.

    https://doi.org/10.1176/appi.ajp.2018.18040369

    • PubMed
    • Google Scholar
57. 1. Scheel AM
    2. Schijen MR
    3. Lakens D

    (2021) An Excess of Positive Results: Comparing the Standard Psychology Literature With Registered Reports

    Advances in Methods and Practices in Psychological Science 4:251524592110074.

    https://doi.org/10.1177/25152459211007467

    • Google Scholar
58. 1. Schumann G
    2. Loth E
    3. Banaschewski T
    4. Barbot A
    5. Barker G
    6. Büchel C
    7. Conrod PJ
    8. Dalley JW
    9. Flor H
    10. Gallinat J
    11. Garavan H
    12. Heinz A
    13. Itterman B
    14. Lathrop M
    15. Mallik C
    16. Mann K
    17. Martinot JL
    18. Paus T
    19. Poline JB
    20. Robbins TW
    21. Rietschel M
    22. Reed L
    23. Smolka M
    24. Spanagel R
    25. Speiser C
    26. Stephens DN
    27. Ströhle A
    28. Struve M
    29. IMAGEN consortium

    (2010) The IMAGEN study: reinforcement-related behaviour in normal brain function and psychopathology

    Molecular Psychiatry 15:1128–1139.

    https://doi.org/10.1038/mp.2010.4

    • PubMed
    • Google Scholar
59. 1. Seo S
    2. Beck A
    3. Matthis C
    4. Genauck A
    5. Banaschewski T
    6. Bokde ALW
    7. Bromberg U
    8. Büchel C
    9. Quinlan EB
    10. Flor H
    11. Frouin V
    12. Garavan H
    13. Gowland P
    14. Ittermann B
    15. Martinot J-L
    16. Paillère Martinot M-L
    17. Nees F
    18. Papadopoulos Orfanos D
    19. Poustka L
    20. Hohmann S
    21. Fröhner JH
    22. Smolka MN
    23. Walter H
    24. Whelan R
    25. Desrivières S
    26. Heinz A
    27. Schumann G
    28. Obermayer K

    (2019) Risk profiles for heavy drinking in adolescence: differential effects of gender

    Addiction Biology 24:787–801.

    https://doi.org/10.1111/adb.12636

    • PubMed
    • Google Scholar
60. 1. Shiraishi J
    2. Li Q
    3. Appelbaum D
    4. Doi K

    (2011) Computer-aided diagnosis and artificial intelligence in clinical imaging

    Seminars in Nuclear Medicine 41:449–462.

    https://doi.org/10.1053/j.semnuclmed.2011.06.004

    • PubMed
    • Google Scholar
61. 1. Smith SM
    2. Jenkinson M
    3. Johansen-Berg H
    4. Rueckert D
    5. Nichols TE
    6. Mackay CE
    7. Watkins KE
    8. Ciccarelli O
    9. Cader MZ
    10. Matthews PM
    11. Behrens TEJ

    (2006) Tract-based spatial statistics: voxelwise analysis of multi-subject diffusion data

    NeuroImage 31:1487–1505.

    https://doi.org/10.1016/j.neuroimage.2006.02.024

    • PubMed
    • Google Scholar
62. 1. Snoek L
    2. Miletić S
    3. Scholte HS

    (2019) How to control for confounds in decoding analyses of neuroimaging data

    NeuroImage 184:741–760.

    https://doi.org/10.1016/j.neuroimage.2018.09.074

    • PubMed
    • Google Scholar
63. 1. Squeglia LM
    2. Schweinsburg AD
    3. Wetherill RR
    4. Pulido C
    5. Tapert SF
    6. Scott F Sorg

    (2012) Binge drinking differentially affects adolescent male and female brain morphometry

    Psychopharmacology 220:529–539.

    https://doi.org/10.1007/s00213-011-2500-4

    • PubMed
    • Google Scholar
64. 1. Squeglia LM
    2. Tapert SF
    3. Sullivan EV
    4. Jacobus J
    5. Meloy MJ
    6. Rohlfing T
    7. Pfefferbaum A

    (2015) Brain development in heavy-drinking adolescents

    The American Journal of Psychiatry 172:531–542.

    https://doi.org/10.1176/appi.ajp.2015.14101249

    • PubMed
    • Google Scholar
65. 1. Squeglia LM
    2. Ball TM
    3. Jacobus J
    4. Brumback T
    5. McKenna BS
    6. Nguyen-Louie TT
    7. Sorg SF
    8. Paulus MP
    9. Tapert SF

    (2017) Neural Predictors of Initiating Alcohol Use During Adolescence

    The American Journal of Psychiatry 174:172–185.

    https://doi.org/10.1176/appi.ajp.2016.15121587

    • PubMed
    • Google Scholar
66. 1. Sudlow C
    2. Gallacher J
    3. Allen N
    4. Beral V
    5. Burton P
    6. Danesh J
    7. Downey P
    8. Elliott P
    9. Green J
    10. Landray M
    11. Liu B
    12. Matthews P
    13. Ong G
    14. Pell J
    15. Silman A
    16. Young A
    17. Sprosen T
    18. Peakman T
    19. Collins R

    (2015) UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age

    PLOS Medicine 12:e1001779.

    https://doi.org/10.1371/journal.pmed.1001779

    • PubMed
    • Google Scholar
67. 1. Sullivan EV
    2. Brumback T
    3. Tapert SF
    4. Brown SA
    5. Baker FC
    6. Colrain IM
    7. Prouty D
    8. De Bellis MD
    9. Clark DB
    10. Nagel BJ
    11. Pohl KM
    12. Pfefferbaum A

    (2020) Disturbed Cerebellar Growth Trajectories in Adolescents Who Initiate Alcohol Drinking

    Biological Psychiatry 87:632–644.

    https://doi.org/10.1016/j.biopsych.2019.08.026

    • PubMed
    • Google Scholar
68. 1. Urbanowicz RJ
    2. Moore JH

    (2015) ExSTraCS 2.0: Description and Evaluation of a Scalable Learning Classifier System

    Evolutionary Intelligence 8:89–116.

    https://doi.org/10.1007/s12065-015-0128-8

    • PubMed
    • Google Scholar
69. 1. Vul E
    2. Harris C
    3. Winkielman P
    4. Pashler H

    (2009) Puzzlingly High Correlations in fMRI Studies of Emotion, Personality, and Social Cognition

    Perspectives on Psychological Science 4:274–290.

    https://doi.org/10.1111/j.1745-6924.2009.01125.x

    • PubMed
    • Google Scholar
70. 1. Wen J
    2. Thibeau-Sutre E
    3. Diaz-Melo M
    4. Samper-González J
    5. Routier A
    6. Bottani S
    7. Dormont D
    8. Durrleman S
    9. Burgos N
    10. Colliot O
    11. Australian Imaging Biomarkers and Lifestyle flagship study of ageing

    (2020) Convolutional neural networks for classification of Alzheimer’s disease: Overview and reproducible evaluation

    Medical Image Analysis 63:101694.

    https://doi.org/10.1016/j.media.2020.101694

    • PubMed
    • Google Scholar
71. 1. Whelan R
    2. Watts R
    3. Orr CA
    4. Althoff RR
    5. Artiges E
    6. Banaschewski T
    7. Barker GJ
    8. Bokde ALW
    9. Büchel C
    10. Carvalho FM
    11. Conrod PJ
    12. Flor H
    13. Fauth-Bühler M
    14. Frouin V
    15. Gallinat J
    16. Gan G
    17. Gowland P
    18. Heinz A
    19. Ittermann B
    20. Lawrence C
    21. Mann K
    22. Martinot J-L
    23. Nees F
    24. Ortiz N
    25. Paillère-Martinot M-L
    26. Paus T
    27. Pausova Z
    28. Rietschel M
    29. Robbins TW
    30. Smolka MN
    31. Ströhle A
    32. Schumann G
    33. Garavan H
    34. IMAGEN Consortium

    (2014) Neuropsychosocial profiles of current and future adolescent alcohol misusers

    Nature 512:185–189.

    https://doi.org/10.1038/nature13402

    • PubMed
    • Google Scholar
72. 1. Wolpert DH
    2. Macready WG

    (1997) No free lunch theorems for optimization

    IEEE Transactions on Evolutionary Computation 1:67–82.

    https://doi.org/10.1109/4235.585893

    • Google Scholar
73. 1. Wyman BT
    2. Harvey DJ
    3. Crawford K
    4. Bernstein MA
    5. Carmichael O
    6. Cole PE
    7. Crane PK
    8. DeCarli C
    9. Fox NC
    10. Gunter JL
    11. Hill D
    12. Killiany RJ
    13. Pachai C
    14. Schwarz AJ
    15. Schuff N
    16. Senjem ML
    17. Suhy J
    18. Thompson PM
    19. Weiner M
    20. Jack CR
    21. Alzheimer’s Disease Neuroimaging Initiative

    (2013) Standardization of analysis sets for reporting results from ADNI MRI data

    Alzheimer’s & Dementia 9:332–337.

    https://doi.org/10.1016/j.jalz.2012.06.004

    • PubMed
    • Google Scholar
74. 1. Yip S
    2. Ye F
    3. Liang Q
    4. Lichenstein S
    5. Dagher A
    6. Pearlson G
    7. Charani B
    8. Garavan H
    9. Scheinost D

    (2022) Neuromarkers of Risky Alcohol Use From Age 14 to 19 Years

    Biological Psychiatry 91:S41–S42.

    https://doi.org/10.1016/j.biopsych.2022.02.122

    • Google Scholar
75. 1. Zahr NM
    2. Pfefferbaum A

    (2017)

    Alcohol’s Effects on the Brain: Neuroimaging Results in Humans and Animal Models

    Alcohol Research 38:183–206.

    • PubMed
    • Google Scholar

Decision letter after peer review:

Thank you for submitting your article "Structural differences in adolescent brains can predict alcohol misuse" for consideration by eLife. Your article has been reviewed by 2 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Chris Baker as the Senior Editor. The reviewers have opted to remain anonymous.

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

Please pay particular attention to the below points raised by one of the reviewers, which I believe to be important.

1) Provide a more convincing explanation for the poor results of the leave-one-site-out predictions.

2) Model longitudinal changes using pre-onset data.

Reviewer #1 (Recommendations for the authors):

This study uses a nice longitudinal dataset and performs relatively thorough methodological comparisons. I also appreciate the systematic literature review presented in the introduction. The discussion of confound control is interesting and it is great that a leave-one-site-out test was included. However, the prediction accuracy drops in these important leave-one-site-out analyses, which should be assessed and discussed further. Furthermore, I think there is a missed opportunity to test longitudinal prediction using only pre-onset individuals to gain clearer causal insights. Please find specific comments below, approximately in order of importance.

1. The leave-one-site-out results fail to achieve significant prediction accuracy for any of the phenotypes. This reveals a lack of cross-site generalizability of all results in this work. The authors discuss that this variance could be caused by distributed sample sizes across sites resulting in uneven folds or site-specific variance. It should be possible to test these hypotheses by looking at the relative performance across CV folds. The site-specific variance hypothesis may be likely because for the other results confounds are addressed using oversampling (i.e., sampling with replacement) which creates a large sample with lower variance than a random sample of the same size. This is an important null finding that may have important implications, so I do not think that it is cause for rejection. However, it is a key element of this paper and I think it should be assessed further and discussed more widely in the abstract and conclusion.

2. The authors state that "83.3% of subjects reported having no or just one binge drinking experience until age 14". To gain clearer insights into the causality, I recommend repeating the MRIage14 → AAMage22 prediction using only these 83% of subjects.

3. The feature importance results for brain regions are quite inconsistent across time points. As such, the study doesn't really address one of the main challenges with previous work discussed in the introduction: "brain regions reported were not consistent between these studies either and do not tell a coherent story". This would be worth looking into further, for example by looking at other indices of feature importance such as permutation-based measures and/or investigating the stability of feature importance across bootstrapped CV folds.

Reviewer #2 (Recommendations for the authors):

I only have a few suggestions.

1) The abstract should be more specific, pointing out the best-predicted measure(binge) and the conclusion of methods, like the conclusion did. Also, the introduction should include more recent prediction studies using IMAGEN.

2) The structure of brain mapping needs clarification. We would like to see the most contributing structural measures（thickness, volume, intensity）separately mapping, how about the accuracy using each type alone? Could you show the brain mapping of each type of 3 sMRI features respectively? Figure 7 is too general.

3) How about the availability of data and code?

Reviewer #1 (Recommendations for the authors):

This study uses a nice longitudinal dataset and performs relatively thorough methodological comparisons. I also appreciate the systematic literature review presented in the introduction. The discussion of confound control is interesting and it is great that a leave-one-site-out test was included. However, the prediction accuracy drops in these important leave-one-site-out analyses, which should be assessed and discussed further. Furthermore, I think there is a missed opportunity to test longitudinal prediction using only pre-onset individuals to gain clearer causal insights. Please find specific comments below, approximately in order of importance.

We thank the reviewers for their positive remarks and for providing important suggestions to improve the analysis. Please see our detailed comments below.

1. The leave-one-site-out results fail to achieve significant prediction accuracy for any of the phenotypes. This reveals a lack of cross-site generalizability of all results in this work. The authors discuss that this variance could be caused by distributed sample sizes across sites resulting in uneven folds or site-specific variance. It should be possible to test these hypotheses by looking at the relative performance across CV folds. The site-specific variance hypothesis may be likely because for the other results confounds are addressed using oversampling (i.e., sampling with replacement) which creates a large sample with lower variance than a random sample of the same size. This is an important null finding that may have important implications, so I do not think that it is cause for rejection. However, it is a key element of this paper and I think it should be assessed further and discussed more widely in the abstract and conclusion.

We thank the reviewer for raising this point and providing specific suggestions. As mentioned by the reviewer, the leave-one-site-out results showed high-variance across sites, that is, across cross validation (CV) folds. Therefore, as suggested by the reviewer, we further investigated the source of this variance by observing how the model accuracies correlates with each site and its sample sizes, ratio of AAM-to-controls, and the sex distribution in each site. We ranked the sites from low to high accuracy and observed different performance metrics such as sensitivity and specificity:

As shown in Appendix—figure 3, the models performed close-to-chance for sites ‘Dublin’, ‘Paris’ and ‘Berlin’ (<60% mean balanced accuracy) in the leave-one-site-out experiment, across all time-points and metrics. Notably, the order of the performance at each site does not correspond to the sample sizes (please refer to the ‘counts’ column in the above figure). It also does not correspond to the ratio of AAM-to-controls, or to the sex distribution.

To further investigate this, we performed another additional leave-one-site-out experiment with all 8 sites. Here, we repeated the ML (Machine Learning) exploration by using the entire data, including the data from the Nottingham site that was kept aside as the holdout. Since there are 8 sites now, we used a 8-fold cross validation and observed how the model accuracy varied across each site:

Author response image 1

Download asset Open asset

The results were comparable to the original leave-one-site-out experiment. Along with ‘Dublin’ and Berlin’, the models additionally performed poorly on the ‘Nottingham’ site. Results on ‘London’ and ‘Paris’ also fell below 60% mean balanced accuracy.Finally, we compared the above two results to the main experiment from the paper where the test samples were randomly sampled across all sites. The performance on test subjects from each site was compared:

As seen in Appendix—figure 2, the models struggled with subjects from ‘Dublin’ followed by ‘Nottingham’ ‘London’ and ‘Berlin’ respectively, and performed well on subjects from ‘Dresden’, ‘Mannheim’, ‘Hamburg’ and ‘Paris’.

Across all the three results discussed above, the models consistently struggle to generalize to subjects particularly from ‘Dublin’ and ‘Nottingham’. As already pointed out by the reviewer, the variance in the main experiment in the manuscript is lower because of the random sampling of the test set across all sites. Since these results have important implications, we have included them in the manuscript and also provided these figures in the Appendix.

2. The authors state that "83.3% of subjects reported having no or just one binge drinking experience until age 14". To gain clearer insights into the causality, I recommend repeating the MRIage14 → AAMage22 prediction using only these 83% of subjects.

3. The feature importance results for brain regions are quite inconsistent across time points. As such, the study doesn't really address one of the main challenges with previous work discussed in the introduction: "brain regions reported were not consistent between these studies either and do not tell a coherent story". This would be worth looking into further, for example by looking at other indices of feature importance such as permutation-based measures and/or investigating the stability of feature importance across bootstrapped CV folds.

Reviewer #2 (Recommendations for the authors):

I only have a few suggestions.

1) The abstract should be more specific, pointing out the best-predicted measure(binge) and the conclusion of methods, like the conclusion did. Also, the introduction should include more recent prediction studies using IMAGEN.

Thank you for the valuable feedback. We agree with these suggestions. We have now revised the abstract to include these important points and have also included more recent IMAGEN studies in the introduction, as shown below. Not all newly added studies are included in Table 1 because this table specifically enlists studies that look at AAM (adolescent alcohol misuse) and use structural MRI data.

Revised Abstract:

“Alcohol misuse during adolescence (AAM) has been associated with disruptive development of adolescent brains. In this longitudinal machine learning (ML) study, we could predict AAM significantly from brain structure (T1-weighted imaging and DTI) with accuracies of 73 – 78% in the IMAGEN dataset (n ~1182). Our results not only show that structural differences in brain can predict AAM, but also suggests that such differences might precede AAM behavior in the data. We predicted ten phenotypes of AAM at age 22 using brain MRI features at ages 14, 19, and 22. Binge drinking was found to be the most predictable phenotype. The most informative brain features were located in the ventricular CSF, and in white matter tracts of the corpus callosum, internal capsule, and brain stem. In the cortex, they were spread across the occipital, frontal, and temporal lobes and in the cingulate cortex. We also experimented with four different ML models and several confound control techniques. Support Vector Machine (SVM) with rbf kernel and Gradient Boosting consistently performed better than the linear models, linear SVM and Logistic Regression. Our study also demonstrates how the choice of the predicted phenotype, ML model, and confound correction technique are all crucial decisions in an explorative ML study analyzing psychiatric disorders with small effect sizes such as AAM.”

New IMAGEN citations:

35. Irina Filippi, Andre´ Galinowski, Herve´ Lemaˆıtre, Christian Massot, Pascal Zille, Pauline Fre`re, Rube´n Miranda-Marcos, Christian Trichard, Stella Guldner, He´le`ne Vulser, et al. Neuroimaging evidence for structural correlates in adolescents resilient to polysubstance use: A five-year follow-up study. European Neuropsychopharmacology, 49:11–22, 2021.

36. Tianye Jia, Chao Xie, Tobias Banaschewski, Gareth J Barker, Arun LW Bokde, Christian Bu¨chel, Erin Burke Quinlan, Sylvane Desrivie`res, Herta Flor, Antoine Grigis, et al. Neural network involving medial orbitofrontal cortex and dorsal periaqueductal gray regulation in human alcohol abuse. Science Advances, 7(6):eabd4074, 2021.

37. Sarah Yip, Fengdan Ye, Qinghao Liang, Sarah Lichenstein, Alecia Dagher, Godfrey Pearlson, Bader Charani, Hugh Garavan, and Dustin Scheinost. Neuromarkers of risky alcohol use from age 14 to 19 years. Biological Psychiatry, 91(9):S41–S42, 2022.

2) The structure of brain mapping needs clarification. We would like to see the most contributing structural measures（thickness, volume, intensity）separately mapping, how about the accuracy using each type alone? Could you show the brain mapping of each type of 3 sMRI features respectively? Figure 7 is too general.

The contributions of different structural feature types was analysed initially during the ML exploration stage, using the 'Binge growth' phenotype. Author response image 2 shows the results from this analysis. Using a single feature type such as only thickness, only volume, or only area resulted in an accuracy that is 6-9% lower than when all features were used together. It suggests that when all features are provided together, the ML models can additionally exploit certain mutual (multivariate) information, that might be present across these different feature types. Therefore, we decided to use all features as input rather than individual feature types and then observe the feature importance post-hoc.

Author response image 2

Download asset Open asset

Nevertheless, we have now provided a table consisting of all the features and feature types with their respective SHAP scores across seven repetitions in the Appendix. This table is aimed to enable future researchers to further investigate the relative contributions of different features and feature types.

3) How about the availability of data and code?

This information has now been provided explicitly in the Data Availability section.

Author details

1. Roshan Prakash Rane

   Charité – Universitätsmedizin Berlin (corporate member of Freie Universiät at Berlin, Humboldt-Universiät at zu Berlin, and Berlin Institute of Health), Department of Psychiatry and Psychotherapy, Bernstein Center for Computational Neuroscience, Berlin, Germany

   Contribution

   Conceptualization, Formal analysis, Methodology, Project administration, Validation, Visualization, Writing - original draft, Writing – review and editing

   For correspondence

   roshan.rane@bccn-berlin.de

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-3996-2034

2. Evert Ferdinand de Man

   Faculty IV – Electrical Engineering and Computer Science, Technische Universität Berlin, Berlin, Germany

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Visualization, Writing - original draft

   Competing interests

   No competing interests declared

3. JiHoon Kim

   Department of Education and Psychology, Freie Universität Berlin, Berlin, Germany

   Contribution

   Investigation, Visualization, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-3157-3472

4. Kai Görgen

   1. Charité – Universitätsmedizin Berlin (corporate member of Freie Universiät at Berlin, Humboldt-Universiät at zu Berlin, and Berlin Institute of Health), Department of Psychiatry and Psychotherapy, Bernstein Center for Computational Neuroscience, Berlin, Germany
   2. Science of Intelligence, Research Cluster of Excellence, Berlin, Germany

   Contribution

   Investigation, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-4711-9629

5. Mira Tschorn

   Social and Preventive Medicine, Department of Sports and Health Sciences, Intra-faculty unit “Cognitive Sciences”, Faculty of Human Science, and Faculty of Health Sciences Brandenburg, Research Area Services Research and e-Health, University of Potsdam, Potsdam, Germany

   Contribution

   Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

6. Michael A Rapp

   Social and Preventive Medicine, Department of Sports and Health Sciences, Intra-faculty unit “Cognitive Sciences”, Faculty of Human Science, and Faculty of Health Sciences Brandenburg, Research Area Services Research and e-Health, University of Potsdam, Potsdam, Germany

   Contribution

   Methodology

   Competing interests

   No competing interests declared

7. Tobias Banaschewski

   Department of Child and Adolescent Psychiatry and Psychotherapy, Central Institute of Mental Health, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany

   Contribution

   Resources

   Competing interests

   No competing interests declared

8. Arun LW Bokde

   Discipline of Psychiatry, School of Medicine and Trinity College Institute of Neuroscience, Trinity College Dublin, Dublin, Ireland

   Contribution

   Data curation, Resources

   Competing interests

   No competing interests declared

9. Sylvane Desrivieres

   Centre for Population Neuroscience and Precision Medicine (PONS), Institute of Psychiatry, Psychology Neuroscience SGDP Centre, King’s College London, London, United Kingdom

   Contribution

   Data curation, Resources

   Competing interests

   No competing interests declared

10. Herta Flor

    1. Institute of Cognitive and Clinical Neuroscience, Central Institute of Mental Health, Medical Faculty Mannheim, Heidelberg University, Heidelberg, Germany
    2. Department of Psychology, School of Social Sciences, University of Mannheim, Mannheim, Germany

    Contribution

    Data curation, Resources, Writing – review and editing

    Competing interests

    No competing interests declared

11. Antoine Grigis

    NeuroSpin, CEA, Université Paris-Saclay, Paris, France

    Contribution

    Data curation, Resources

    Competing interests

    No competing interests declared

12. Hugh Garavan

    Departments of Psychiatry and Psychology, University of Vermont, Burlington, United States

    Contribution

    Data curation, Resources

    Competing interests

    No competing interests declared

13. Penny A Gowland

    Sir Peter Mansfield Imaging Centre School of Physics and Astronomy, University of Nottingham, Nottingham, United Kingdom

    Contribution

    Data curation, Resources

    Competing interests

    No competing interests declared

14. Rüdiger Brühl

    Physikalisch-Technische Bundesanstalt, Berlin, Germany

    Contribution

    Data curation, Resources

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0003-0111-5996

15. Jean-Luc Martinot

    Institut National de la Santé et de la Recherche Médicale, INSERM U A10 ”Trajectoires développementales en psychiatrie” Universite Paris-Saclay, Ecole Normale Supérieure Paris-Saclay, CNRS, Centre Borelli, Gif-sur-Yvette, France

    Contribution

    Data curation, Resources

    Competing interests

    No competing interests declared

16. Marie-Laure Paillere Martinot

    1. Institut National de la Santé et de la Recherche Médicale, INSERM U A10 ”Trajectoires développementales en psychiatrie” Universite Paris-Saclay, Ecole Normale Supérieure Paris-Saclay, CNRS, Centre Borelli, Gif-sur-Yvette, France
    2. AP-HP Sorbonne Université, Department of Child and Adolescent Psychiatry, Pitié-Salpêtrière Hospital, Paris, France

    Contribution

    Data curation, Resources

    Competing interests

    No competing interests declared

17. Eric Artiges

    1. Institut National de la Santé et de la Recherche Médicale, INSERM U A10 ”Trajectoires développementales en psychiatrie” Universite Paris-Saclay, Ecole Normale Supérieure Paris-Saclay, CNRS, Centre Borelli, Gif-sur-Yvette, France
    2. Psychiatry Department, EPS Barthélémy Durand, Etampes, France

    Contribution

    Data curation, Resources

    Competing interests

    No competing interests declared

18. Frauke Nees

    1. Department of Child and Adolescent Psychiatry and Psychotherapy, Central Institute of Mental Health, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany
    2. Institute of Cognitive and Clinical Neuroscience, Central Institute of Mental Health, Medical Faculty Mannheim, Heidelberg University, Heidelberg, Germany
    3. PONS Research Group, Dept of Psychiatry and Psychotherapy, Campus Charite Mitte, Humboldt University, Berlin, Germany

    Contribution

    Data curation, Resources

    Competing interests

    No competing interests declared

19. Dimitri Papadopoulos Orfanos

    NeuroSpin, CEA, Université Paris-Saclay, Paris, France

    Contribution

    Data curation, Resources

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0002-1242-8990

20. Herve Lemaitre

    1. NeuroSpin, CEA, Université Paris-Saclay, Paris, France
    2. Institut des Maladies Neurodégénératives, UMR 5293, CNRS, CEA, University of Bordeaux, Bordeaux, France

    Contribution

    Data curation, Resources, Writing – review and editing

    Competing interests

    No competing interests declared

21. Tomas Paus

    1. Department of Psychiatry, Faculty of Medicine and Centre Hospitalier Universitaire Sainte-Justine, University of Montreal, Montreal, Canada
    2. Departments of Psychiatry and Psychology, University of Toronto, Toronto, Canada

    Contribution

    Data curation, Resources

    Competing interests

    No competing interests declared

22. Luise Poustka

    Department of Child and Adolescent Psychiatry and Psychotherapy, University Medical Centre Göttingen, Göttingen, Germany

    Contribution

    Data curation, Resources

    Competing interests

    No competing interests declared

23. Juliane Fröhner

    Department of Psychiatry and Neuroimaging Center, Technische Universität Dresden, Dresden, Germany

    Contribution

    Data curation, Resources

    Competing interests

    No competing interests declared

24. Lauren Robinson

    Department of Psychological Medicine, Section for Eating Disorders, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, London, United Kingdom

    Contribution

    Resources

    Competing interests

    No competing interests declared

25. Michael N Smolka

    Department of Psychiatry and Neuroimaging Center, Technische Universität Dresden, Dresden, Germany

    Contribution

    Resources

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0001-5398-5569

26. Jeanne Winterer

    1. Charité – Universitätsmedizin Berlin (corporate member of Freie Universiät at Berlin, Humboldt-Universiät at zu Berlin, and Berlin Institute of Health), Department of Psychiatry and Psychotherapy, Bernstein Center for Computational Neuroscience, Berlin, Germany
    2. Department of Education and Psychology, Freie Universität Berlin, Berlin, Germany

    Contribution

    Resources

    Competing interests

    No competing interests declared

27. Robert Whelan

    School of Psychology and Global Brain Health Institute, Trinity College Dublin, Dublin, Ireland

    Contribution

    Resources

    Competing interests

    No competing interests declared

28. Gunter Schumann

    PONS Research Group, Dept of Psychiatry and Psychotherapy, Campus Charite Mitte, Humboldt University, Berlin, Germany

    Contribution

    Data curation, Resources, Writing – review and editing

    Competing interests

    No competing interests declared

29. Henrik Walter

    Charité – Universitätsmedizin Berlin (corporate member of Freie Universiät at Berlin, Humboldt-Universiät at zu Berlin, and Berlin Institute of Health), Department of Psychiatry and Psychotherapy, Bernstein Center for Computational Neuroscience, Berlin, Germany

    Contribution

    Project administration, Writing – review and editing

    Competing interests

    No competing interests declared

30. Andreas Heinz

    Charité – Universitätsmedizin Berlin (corporate member of Freie Universiät at Berlin, Humboldt-Universiät at zu Berlin, and Berlin Institute of Health), Department of Psychiatry and Psychotherapy, Bernstein Center for Computational Neuroscience, Berlin, Germany

    Contribution

    Funding acquisition, Investigation, Writing – review and editing

    Competing interests

    No competing interests declared

31. Kerstin Ritter

    Charité – Universitätsmedizin Berlin (corporate member of Freie Universiät at Berlin, Humboldt-Universiät at zu Berlin, and Berlin Institute of Health), Department of Psychiatry and Psychotherapy, Bernstein Center for Computational Neuroscience, Berlin, Germany

    Contribution

    Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing - original draft, Writing – review and editing

    Competing interests

    No competing interests declared

32. IMAGEN consortium

    Contribution

    Data curation, Resources

    Competing interests

    No competing interests declared