Expansion of the Frontostriatal Salience Network in Depression: A Precision Functional Mapping Study

Table of Contents

Overall Summary

Overview

This study investigated the topology of the frontostriatal salience network in individuals with depression using precision functional mapping and longitudinal fMRI data. The researchers found that the salience network is significantly expanded in the cortex of most individuals with depression, a finding replicated across multiple independent samples and datasets. This expansion appears to be a stable, trait-like feature, detectable even before depression onset in children, and may be related to specific patterns of encroachment on neighboring brain networks.

Key Findings

Strengths

Areas for Improvement

Significant Elements

Figure 1

Description: Illustrates the significant expansion of the salience network in the cortex of individuals with depression compared to healthy controls, providing visual and quantitative evidence of the effect size and replicability across samples.

Relevance: Provides the core visual evidence for the primary finding of the study.

Figure 2

Description: Demonstrates the heterogeneity of salience network expansion by identifying three distinct encroachment profiles affecting neighboring functional systems, highlighting individual variability in how the network is altered in depression.

Relevance: Expands upon the primary finding by revealing distinct patterns of network reorganization.

Conclusion

This study provides compelling evidence for a robust and reproducible expansion of the frontostriatal salience network in individuals with depression. The findings suggest that this expansion may be a stable, trait-like marker for depression risk, potentially detectable before symptom onset. Further research is needed to explore the causal relationships, functional consequences, and specificity of this finding, but it holds promise for advancing our understanding of depression and developing new strategies for early detection and personalized interventions.

Section Analysis

Abstract

Overview

This abstract presents a study that used precision functional mapping to investigate brain network topology in individuals with depression. The study found that the frontostriatal salience network is significantly expanded in the cortex of most individuals with depression, a finding that was replicated across multiple samples. This expansion appears to be a stable, trait-like feature present even before the onset of depression in children. The study also found that changes in connectivity within the salience network, particularly between the striatum and anterior cingulate cortex, tracked fluctuations in anhedonia symptoms and predicted future anhedonia.

Key Aspects

Strengths

Suggestions for Improvement

Introduction

Overview

This introduction section establishes the context for a study investigating the topology of functional brain networks in individuals with depression, particularly focusing on the salience network. It highlights the limitations of previous neuroimaging studies in depression, which have primarily relied on cross-sectional designs and group-average parcellations, resulting in modest effect sizes and limited understanding of mood-state transitions. The section introduces the concept of precision functional mapping as a novel approach to delineate individual-specific network topology and its potential to reveal more meaningful differences in brain organization related to depression.

Key Aspects

Strengths

Suggestions for Improvement

Results

Overview

This Results section presents the findings of a study investigating the topology of the frontostriatal salience network in individuals with depression. The key finding is a significant expansion of this network in the cortex of depressed individuals, a phenomenon replicated across multiple samples and datasets. The section further explores the nature of this expansion, identifying three distinct modes of encroachment on neighboring networks, and investigates its stability over time, relationship to symptom severity, and presence before symptom onset in children. Finally, the section examines the dynamic changes in functional connectivity within the salience network, particularly between the striatum and anterior cingulate cortex, and their relationship to fluctuations in anhedonia symptoms.

Key Aspects

Strengths

Suggestions for Improvement

Non-Text Elements

Figure Figure 1

Figure 1, titled 'Frontostriatal salience network is expanded nearly twofold in the cortex of highly sampled individuals with depression,' illustrates the expansion of the salience network in individuals with depression compared to healthy controls. The figure includes brain images showing the location and size of the salience network, scatter plots comparing the percentage of cortex occupied by the network in different groups (73% larger in the SIMD dataset), density maps highlighting the expansion in cortical zones, and a confusion matrix demonstrating the accuracy (78.4%) of an SVM classifier in distinguishing individuals with depression from healthy controls based on functional network sizes. Key numeric data includes: 5.49% ± 0.76% of cortex occupied by the salience network in the SIMD dataset, 3.17% ± 0.85% in healthy controls, Cohen's d of 1.99 for the group difference, and classifier accuracy of 78.4%.

First Mention

Text: "It was immediately apparent on visual inspection that the salience network, which is involved in reward processing and conscious integration of autonomic feedback and responses with internal goals and environmental demands30,40,41, was markedly larger in these individuals with depression (Fig. 1a,b)."

Context: This sentence appears in the third paragraph of the 'Salience network expansion in depression' section on page 3. It introduces the observation that the salience network is visibly larger in individuals with depression, as depicted in Figure 1.

Relevance: This figure is central to the section's main point about the expanded salience network in depression. It provides visual and quantitative evidence supporting this claim, demonstrating the effect's magnitude and replicability across different samples and methodologies.

Critique
Visual Aspects
  • The figure is generally clear and well-organized, with distinct color schemes and labels for different elements.
  • The use of multiple sub-elements (brain images, scatter plots, density maps, confusion matrix) effectively conveys different aspects of the research.
  • The caption is extremely long and detailed, making it challenging to follow. Some of this information could be integrated into the main text for better readability.
Analytical Aspects
  • The figure incorporates a robust set of statistical methods, including permutation tests, t-tests, SVM classification, and feature importance analysis, to support the findings.
  • The inclusion of replication datasets strengthens the reliability of the observed effect.
  • The analysis focuses primarily on cortical expansion, with limited exploration of potential differences in subcortical regions, which could be further investigated.
Numeric Data
  • Average salience network size in SIMD dataset: 5.49 % of cortex
  • Average salience network size in healthy controls: 3.17 % of cortex
  • Effect size (Cohen's d) for group difference: 1.99
  • SVM classifier accuracy: 78.4 %
  • Positive predictive value of SVM classifier: 89.5 %
Figure Fig. 2

Figure 2, titled 'Three modes of salience network expansion in depression,' explores the different patterns of salience network expansion in individuals with depression. It shows a central tendency functional network map for healthy controls, identifies encroaching parts of the salience network in individuals with depression, and classifies them as ectopic intrusions or border shifts. The figure also presents bar graphs quantifying the frequency of border shifts and ectopic intrusions, the networks most often displaced by salience network expansion (default mode, frontoparietal, and cingulo-opercular networks), and the network displacement in different cortical zones. A heatmap reveals three distinct modes of encroachment across individuals.

First Mention

Text: "To this end, we first generated a central tendency functional network map for the 37 healthy controls (Fig. 2a)."

Context: This sentence, found in the second paragraph of the 'Three salience network expansion modes' section on page 4, introduces the first step in analyzing the different modes of salience network expansion, as depicted in Figure 2a.

Relevance: Figure 2 builds upon the findings of Figure 1 by delving into the heterogeneity of salience network expansion. It demonstrates that this expansion is not random but follows specific patterns, primarily involving border shifts affecting neighboring higher-order functional systems.

Critique
Visual Aspects
  • The figure effectively uses color-coding to differentiate various brain networks and their interactions.
  • The bar graphs and heatmap are clear and easy to interpret, providing a good visual representation of the data.
  • The figure could benefit from including brain images with clearer demarcations of the encroaching salience network regions for better visual understanding.
Analytical Aspects
  • The analysis provides valuable insights into the specific functional systems affected by salience network expansion, suggesting a potential reallocation of cortical territory.
  • The identification of three distinct encroachment modes highlights the individual variability in depression and the need for personalized approaches.
  • The study could benefit from further investigation into the functional consequences of these different encroachment modes and their relationship to specific symptom profiles.
Numeric Data
  • Percentage of individuals with salience network expansion driven by border shifts: 75 %
  • Percentage of individuals with salience network expansion driven by ectopic intrusions: 25 %
  • Percentage of encroachment on the default mode network in the anterior cingulate cortex: 60 %
  • Percentage of encroachment on the frontoparietal network in the lateral prefrontal cortex: 60 %
  • Percentage of encroachment on the cingulo-opercular network in the anterior insular cortex: 45 %
Figure Figure 3

Figure 3, titled 'Salience network expansion is stable over time and present before symptom onset,' investigates the stability of salience network expansion and its presence before the onset of depression. It includes line graphs showing the stability of salience network size across multiple study visits in both healthy controls and individuals with depression, scatter plots demonstrating no significant correlation between salience network size and depression severity (HDRS6), symptom severity, or chronicity, and a bar graph indicating no change in salience network size after rTMS treatment. A key finding is that the salience network is larger in children who later developed depression compared to those who did not, suggesting its presence before symptom onset.

First Mention

Text: "Furthermore, within-subject analyses showed no significant correlation between fluctuations in depression symptoms (HDRS6, a more sensitive measure of changes on shorter timescales) and changes in salience network size over time in any of the densely sampled individuals in our SIMD dataset (Fig. 3b)."

Context: This sentence, located in the first paragraph of the 'Salience network topology is trait-like' section on page 4, highlights the lack of correlation between symptom fluctuations and salience network size, as illustrated in Figure 3b.

Relevance: Figure 3 addresses a crucial question about the nature of salience network expansion: is it a state-dependent marker fluctuating with mood or a stable trait-like feature? The figure's data supports the latter, showing the expansion's stability over time and its presence even before symptom onset, suggesting a potential role in depression risk.

Critique
Visual Aspects
  • The figure is well-organized and clearly labeled, with different subplots effectively conveying multiple aspects of the research findings.
  • The color scheme and data representation are appropriate and easy to interpret.
  • The figure could benefit from a clearer visual separation between the different subplots to enhance readability.
Analytical Aspects
  • The longitudinal data and analyses provide strong evidence for the stability of salience network expansion, challenging the notion of it being a state-dependent marker.
  • The finding of salience network expansion in children before depression onset is particularly compelling, suggesting a potential predictive value.
  • The study could benefit from further investigation into the factors that might contribute to salience network expansion early in life, such as genetic predisposition or early life experiences.
Numeric Data
  • Correlation between HDRS6 and salience network size in SIMD individuals: -0.06
  • P-value for the correlation between HDRS6 and salience network size:
  • Change in salience network size after 6-week rTMS:
  • P-value for the change in salience network size after 6-week rTMS: 0.55
  • Change in salience network size after 1-week rTMS:
  • P-value for the change in salience network size after 1-week rTMS: 0.56
Figure Figure 4

Figure 4, titled 'Frontostriatal salience network connectivity predicts fluctuations in anhedonia and anxiety symptoms in deeply sampled individuals with depression over time,' examines the relationship between salience network connectivity and symptom fluctuations. It includes a heatmap showing variations in anhedonia-related symptoms over time, brain images highlighting the relevant network nodes, correlation matrices depicting the association between connectivity strength and symptoms, and scatter plots showing the predictive relationship between connectivity and future symptoms. Key findings include: connectivity between the nucleus accumbens and anterior cingulate cortex tracks anhedonia fluctuations and predicts future anhedonia in one individual, while connectivity between the nucleus accumbens and anterior insula correlates with anxiety fluctuations.

First Mention

Text: "During a period spanning 1.5 years, we observed significant fluctuations in ten anhedonia-related measures (Fig. 4a), which were derived from five standardized depressive symptom scales and identified by a consensus clinical decision by three study co-authors (Supplementary Fig. 9), ranging from mild or negligible to severe."

Context: This sentence, found in the first paragraph of the 'Connectivity state predicts anhedonia' section on page 6, describes the observed fluctuations in anhedonia-related measures in one individual (SIMD-4), as depicted in Figure 4a.

Relevance: Figure 4 shifts the focus from network topology to connectivity, demonstrating that while the salience network's size is stable, its connectivity fluctuates with mood state. This finding suggests that connectivity changes within the salience network, particularly in frontostriatal circuits, might contribute to the dynamic nature of depressive symptoms.

Critique
Visual Aspects
  • The figure is generally well-organized and clear, with each sub-figure addressing a specific aspect of the research question.
  • The color schemes and labels are easy to follow, and the brain images effectively highlight the relevant anatomical regions.
  • The caption is separated from the figure by a large block of text, which could potentially cause confusion for readers.
Analytical Aspects
  • The longitudinal analysis of deeply sampled individuals provides valuable insights into the dynamic relationship between connectivity and symptoms.
  • The finding that connectivity changes predict future anhedonia in one individual is intriguing but requires further validation in larger samples.
  • The study could benefit from exploring the directionality of the observed connectivity changes and investigating potential causal relationships between connectivity and symptoms.
Numeric Data
  • Correlation between nucleus accumbens-anterior cingulate connectivity and anhedonia in SIMD-4: -0.37
  • P-value for the correlation between nucleus accumbens-anterior cingulate connectivity and anhedonia in SIMD-4: 0.003
  • Correlation between nucleus accumbens-anterior cingulate connectivity and anhedonia in SIMD-6: -0.49
  • P-value for the correlation between nucleus accumbens-anterior cingulate connectivity and anhedonia in SIMD-6: 0.001
  • Correlation between nucleus accumbens-anterior cingulate connectivity and future anhedonia in SIMD-4: -0.32
  • P-value for the correlation between nucleus accumbens-anterior cingulate connectivity and future anhedonia in SIMD-4: 0.004
Figure Extended Data Fig. 1 | Serial Imaging of Major Depression.

Extended Data Figure 1 provides a visual overview of the Serial Imaging of Major Depression (SIMD) project. It includes a timeline illustrating the study procedures, which involved repeated multi-echo resting-state fMRI scans and clinical assessments. The figure also depicts brain surface renderings showcasing functional network parcellations and surface area representations. Additionally, it presents a stacked bar chart comparing network sizes between depression and control groups, a line graph visualizing symptom severity fluctuations over time, and a bar chart displaying network size calculations for different brain regions.

First Mention

Text: "The same precision mapping procedures were applied to 37 highly sampled healthy controls with an average of 327.49 min of fMRI data per subject (range 43.36–841.2 min) across 12 sessions (range 2–84 sessions)."

Context: This sentence, appearing early in the 'Salience network expansion in depression' section, introduces the healthy control sample used for comparison and mentions that the same precision mapping procedures were applied to them, as detailed in Extended Data Figure 1.

Relevance: This figure is relevant because it visually outlines the study design and methods of the SIMD project, providing context for the subsequent findings on salience network expansion in individuals with depression. It helps the reader understand the data acquisition process and the approach to analyzing functional network topology.

Critique
Visual Aspects
  • The figure's layout is somewhat cluttered due to the combination of diverse elements, but each component is individually clear and labeled.
  • The use of color in the brain renderings and bar charts enhances data visualization.
  • However, the figure lacks a clear visual hierarchy, making it challenging to discern the primary focus.
Analytical Aspects
  • While the figure presents data comparisons (e.g., network sizes between groups), it doesn't explicitly display statistical significance markers like p-values or confidence intervals within the visual elements.
  • The caption mentions statistical analyses but doesn't specify the methods used.
Figure Extended Data Fig. 2 | Salience network expansion in depression remains statistically significant when controlling for sex ratio imbalance, and individual differences in head motion and age.

Extended Data Figure 2 presents six scatter plots demonstrating that the observed salience network expansion in individuals with depression remains statistically significant even after controlling for potential confounding variables. The plots compare the percentage of cortex occupied by the salience network in healthy controls and individuals with depression, accounting for sex ratio imbalance, head motion, and age. The figure shows that the salience network is consistently larger in individuals with depression across all three replication samples and the discovery sample, regardless of the covariates included.

First Mention

Text: "This effect was replicated thrice (Fig. 1c, right), again with medium to large effect sizes (Cohen’s d = 0.77–0.84), remained statistically significant when controlling for the sex ratio imbalance in our samples (56.7% of individuals with depression were female, versus 31% of the healthy controls; Supplementary Fig. 4 and Extended Data Fig. 2) and with or without correction for potential site- or scanner-induced biases (Supplementary Fig. 5)."

Context: This sentence, following the initial description of salience network expansion in individuals with depression, emphasizes the replication of the finding and mentions that the effect remained statistically significant even after controlling for sex ratio imbalance, as detailed in Extended Data Figure 2.

Relevance: This figure is crucial in addressing potential concerns about confounding variables influencing the observed salience network expansion. It strengthens the study's findings by demonstrating that the effect is not solely driven by differences in sex, head motion, or age between the groups.

Critique
Visual Aspects
  • The figure is well-organized and clearly labeled. The use of separate plots for different covariates and sample types aids in understanding the data.
  • The error bars and asterisks provide clear visual cues for statistical significance.
  • However, the figure caption lacks specific details about the statistical tests conducted.
Analytical Aspects
  • While the figure itself primarily uses asterisks to denote statistical significance, the caption provides further details, mentioning the use of two-tailed independent sample t-tests.
  • The caption reports statistically significant differences and effect sizes for both the discovery and replication samples.
  • However, the figure itself does not visually represent the additional analyses conducted using residuals after regressing salience network size against covariates.
Figure Extended Data Fig. 3 | Expansion of the salience network accompanied by contraction of neighboring functional systems. a, b

Extended Data Figure 3 explores the relationship between salience network expansion and the size of neighboring functional systems. It presents brain images depicting the salience network, default mode network, frontoparietal network, and cingulo-opercular network in both a group-average map of healthy controls and three representative individuals with depression. The figure also includes scatter plots comparing the size of these networks between healthy controls and individuals with depression. Notably, the figure suggests that while salience network expansion is consistently observed, the contraction of other networks, particularly the cingulo-opercular network, is less consistent across samples.

First Mention

Text: "Accordingly, expansion of the salience network in cortex was accompanied by contraction of neighbouring functional systems in the SIMD sample (Extended Data Fig. 3)."

Context: This sentence, appearing after the description of salience network expansion in individuals with depression, introduces the observation that this expansion is accompanied by a contraction of neighboring functional systems, as illustrated in Extended Data Figure 3.

Relevance: This figure is relevant because it investigates the potential consequences of salience network expansion on the organization of other functional brain networks. It suggests that the brain may compensate for the increased salience network size by reducing the size of neighboring networks, although this effect appears to be variable across datasets.

Critique
Visual Aspects
  • The figure is generally clear, with distinct color coding for different brain regions.
  • However, a more detailed legend explaining the colors would enhance clarity.
  • The separation of the caption from the image by a large block of text is somewhat unusual and could be improved for better readability.
Analytical Aspects
  • While the specific statistical methods are not directly visible in the figure, the caption suggests the use of permutation tests, with a p-value of 0.04 mentioned.
  • The caption also notes that error bars represent standard deviation.
  • However, the figure lacks a clear visual representation of the statistical significance of the observed differences in network size between groups.
Figure Extended Data Fig. 4 | Evidence of salience network expansion in large *n* group-average datasets. a, Salience network mapped using two large *n* group-average data from previous studies of healthy controls occupy 1.27% and 1.98% of cortex. The group-average HCP functional connectivity matrix (which only includes subjects with resting-state fMRI data reconstructed with the r227 recon algorithm) was obtained from the S1200 release and subjected to the same precision functional mapping procedures applied to individual subjects in the main text. The WU120 salience network map was obtained online (https://balsa.wustl.edu/jNXKI). b, Salience network mapped using large *n* group-average data and previous studies of depression occupies between 3.28% (mode assignment of all individuals with depression in current study) and 3.43% of total cortical surface area. Group-averaged functional connectivity was calculated in the THREE-D sample using group-level PCA (MELODIC Incremental Group-PCA, MIGP), and the resultant group-average FC matrix was subjected to the same precision functional mapping procedures applied to individual subjects in the main text.

Extended Data Figure 4 provides further evidence for salience network expansion in individuals with depression by analyzing large group-average datasets. It presents brain images depicting the salience network mapped using data from previous studies of both healthy controls and individuals with depression. The figure shows that the salience network occupies a significantly larger percentage of cortex in the depression samples compared to the healthy control samples, supporting the findings from the individual-level analyses.

First Mention

Text: "To better understand whether this effect was also detectable in large, previously published samples involving conventional single-echo fMRI data, we identified the salience network in group-average functional connectivity data from two large datasets involving n = 812 (ref. 47) and n = 120 (ref. 45) healthy controls, respectively, and in a third dataset involving n = 299 individuals with treatment-resistant depression scanned in association with a neuromodulation intervention study48."

Context: This sentence, appearing towards the end of the 'Salience network expansion in depression' section, introduces the analysis of large group-average datasets to further validate the finding of salience network expansion in individuals with depression, as detailed in Extended Data Figure 4.

Relevance: This figure is important because it demonstrates that the observed salience network expansion is not limited to the specific samples analyzed in the study but is also detectable in large, independent datasets. This strengthens the generalizability of the findings and suggests that salience network expansion may be a robust feature of depression.

Critique
Visual Aspects
  • The figure is generally clear and well-organized. The labeling of different brain regions and the use of different views (lateral and medial) effectively communicate the spatial extent of the salience network.
  • The percentages provided offer a quantifiable measure of the network's size.
  • However, the caption is quite lengthy and contains detailed methodological information that might be better suited for a separate methods section.
Analytical Aspects
  • While the figure itself doesn't explicitly depict statistical methods, the caption mentions the use of group-level PCA to calculate group-averaged functional connectivity.
  • This suggests that statistical methods were employed in the data analysis, but the specific details and results are not visually represented in the figure.
  • The figure would benefit from including statistical significance markers or effect sizes to quantify the differences in salience network size between groups.
Figure Extended Data Fig. 5 | Within-person stability of salience network topology and connectivity. a-b, Split-half reliability testing of salience network topology and functional connectivity in the least (SIMD-2, 2.58 min of fMRI

Extended Data Figure 5 assesses the within-person stability of salience network topology and connectivity using split-half reliability testing. It presents brain images depicting the salience network in two individuals with depression (SIMD-2 and SIMD-4), comparing the network maps derived from the first and second halves of their fMRI data. The figure shows highly similar patterns of salience network topology and functional connectivity in both individuals, indicating that the observed expansion is a stable and reproducible feature within individuals.

First Mention

Text: "Furthermore, highly similar patterns of salience network topology and functional connectivity were produced in split-half analyses of each SIMD dataset (Extended Data Fig. 5), indicating that salience network expansion was a robust and reproducible feature of the brains of these highly sampled individuals."

Context: This sentence, appearing at the end of the 'Salience network expansion in depression' section, highlights the stability and reproducibility of salience network expansion within individuals, as demonstrated by split-half analyses detailed in Extended Data Figure 5.

Relevance: This figure is relevant because it addresses the reliability of the observed salience network expansion. By demonstrating high split-half reliability, the figure supports the notion that the expansion is not due to random fluctuations in the data but reflects a stable characteristic of brain organization in individuals with depression.

Critique
Visual Aspects
  • The figure is relatively clear and well-organized, with distinct subfigures and labels for different conditions.
  • However, without further context or a more detailed caption, interpreting the specific findings and the meaning of the color gradients remains challenging.
Analytical Aspects
  • While the figure itself doesn't explicitly show statistical measures, the caption mentions 'split-half reliability testing,' implying that some form of correlation or consistency metric was used.
  • The 'FC' values likely represent these reliability coefficients, but the specific method used to calculate them isn't stated.
  • The figure would benefit from including quantitative measures of reliability, such as correlation coefficients or dice similarity coefficients, to provide a more precise assessment of network stability.
Figure Extended Data Fig. 6

Extended Data Fig. 6 explores the relationship between salience network expansion and other brain characteristics. Panel (a) is a bar chart showing the percentage enrichment of different brain networks affected by salience network expansion. Panel (b) includes a dot plot depicting the similarity of the average salience network encroachment map to 73 canonical brain maps from the neuromaps toolbox, along with brain images visualizing some of these maps. The dot plot highlights significant associations with gradients of functional connectivity and gene expression, neurotransmitter receptor distribution, intracortical myelin, and individual variability in functional connectivity.

First Mention

Text: "Comparison to 73 independent molecular, microstructural, electrophysiological, developmental and functional brain maps from neuromaps toolbox53 showed that salience network expansion frequently occurred in brain regions with less intracortical myelin and thus greater capacity for synaptic plasticity54 and for which individual differences in functional connectivity55 and the concentration of particular neurotransmitter receptors (μ-opioid56 and histamine H3 receptors57) are most pronounced (Extended Data Fig. 6)."

Context: This sentence, located in the 'Three salience network expansion modes' section on page 4, describes the analysis comparing the salience network encroachment map to various brain characteristics using the neuromaps toolbox. It highlights the finding that salience network expansion is associated with specific brain regions characterized by lower intracortical myelin, greater plasticity, and distinct patterns of functional connectivity and neurotransmitter receptor distribution.

Relevance: This figure provides insights into the potential biological mechanisms underlying salience network expansion in depression. It suggests that this expansion is not random but rather targets specific brain regions with distinct molecular, microstructural, and functional properties, potentially reflecting altered plasticity and information processing priorities.

Critique
Visual Aspects
  • The bar chart in panel (a) is clear and easy to interpret, effectively showing the relative enrichment of different networks.
  • The dot plot in panel (b) is informative but could benefit from clearer labeling of the y-axis to improve readability.
  • The brain images in panel (b) are visually appealing and effectively illustrate the spatial distribution of the selected brain maps.
Analytical Aspects
  • The analysis comparing the salience network encroachment map to a wide range of brain characteristics is comprehensive and provides valuable insights into potential mechanisms.
  • The use of spatial autocorrelation preserving null models ensures robust statistical evaluation of the observed associations.
  • The figure would benefit from quantifying the strength of the associations (e.g., correlation coefficients) in addition to indicating statistical significance.
Numeric Data
  • Default-Parietal Network Enrichment: 35 %
  • Frontoparietal Network Enrichment: 30 %
  • Cingulo-opercular Network Enrichment: 25 %
  • Similarity to Intracortical Myelin Map: -0.4
  • Similarity to μ-opioid Receptor Map: 0.3
Figure Extended Data Fig. 7 | See next page for caption.

Extended Data Fig. 7, with its full caption on the following page, investigates the stability of atypical functional connectivity patterns associated with salience network expansion in depression. It presents two heatmaps (a and b) showing the functional connectivity strength between encroaching and non-encroaching vertices of the salience network relative to runner-up network assignments, using split halves of each individual's resting-state fMRI data. The accompanying bar graphs display the average functional connectivity strength for each network. The caption highlights that functional connectivity between encroaching salience network vertices and the rest of the salience network was significantly stronger than with the runner-up networks, indicating stable atypical connectivity patterns.

First Mention

Text: "As expected, the functional connectivity of encroaching salience network nodes with the rest of the salience network was significantly stronger (mean Z(r) = 0.26) than with the displaced networks (all mean Z(r) < 0.12), consistent with weakened connectivity between encroaching nodes and the functional networks that typically occupy that space in healthy controls (Extended Data Fig. 7)."

Context: This sentence, found in the 'Three salience network expansion modes' section on page 4, describes the analysis comparing the functional connectivity strength of encroaching salience network nodes with the rest of the salience network versus the displaced networks. It highlights the finding that encroaching nodes show stronger connectivity with the salience network, supporting the idea of altered network allocation.

Relevance: This figure supports the finding that salience network expansion in depression is not merely a spatial phenomenon but also involves stable alterations in functional connectivity. It demonstrates that encroaching salience network nodes exhibit stronger connectivity with the salience network than with the networks they displace, suggesting a functional reallocation of these regions.

Critique
Visual Aspects
  • The heatmaps provide a clear visualization of the functional connectivity patterns, with the color scale effectively representing connectivity strength.
  • The bar graphs above the heatmaps are helpful in summarizing the average connectivity strength for each network.
  • The figure would benefit from clearer labeling of the brain regions represented in the heatmaps.
Analytical Aspects
  • The use of split-half analyses demonstrates the stability of the observed connectivity patterns.
  • The statistical analysis using t-tests with Bonferroni correction ensures robust evaluation of the group differences.
  • The figure would be strengthened by quantifying the effect size of the connectivity differences.
Numeric Data
  • Average Connectivity Strength (Encroaching Salience - Salience): 0.26 Z(r)
  • Average Connectivity Strength (Encroaching Salience - Runner-up): 0.12 Z(r)
  • Percentage Stronger Connectivity (Encroaching Salience - Salience vs. Runner-up): 59 %
  • Number of Subjects: 141
  • P-value (t-test): 0.001
Figure Extended Data Fig. 7 | Salience network expansion in depression is associated with stable patterns of atypical functional connectivity. a-b

This figure, a continuation of Extended Data Fig. 7, delves deeper into the stability of atypical functional connectivity associated with salience network expansion. It uses split-half analyses of resting-state fMRI data to compare the functional connectivity strength of encroaching salience network vertices with the rest of the salience network versus the networks they typically displace in healthy controls. The figure demonstrates that encroaching vertices exhibit significantly stronger connectivity with the salience network, a pattern consistent across both halves of the data, indicating stability.

First Mention

Text: "This analysis was performed using split halves of each individual’s resting-state fMRI dataset to assess the stability of the salience network assignment associated with the encroaching vertices relative to the runner-up assignments (most often either the default mode, frontoparietal or cingulo-opercular network)."

Context: This sentence, located in the 'Three salience network expansion modes' section on page 4, introduces the split-half analysis used to evaluate the stability of the salience network assignment for encroaching vertices. It emphasizes the comparison to runner-up network assignments, typically the default mode, frontoparietal, or cingulo-opercular networks.

Relevance: This figure reinforces the concept that salience network expansion in depression is accompanied by stable changes in functional connectivity. The split-half analyses provide strong evidence that encroaching salience network vertices are functionally more connected to the salience network than to the networks they displace, supporting the notion of a functional reallocation of cortical territory.

Critique
Visual Aspects
  • The figure effectively uses bar graphs to represent the average functional connectivity strength for different network pairings.
  • The use of separate bars for each half of the data clearly demonstrates the consistency of the findings.
  • The figure would benefit from a clearer visual separation between the two halves of the data (e.g., different colors or patterns).
Analytical Aspects
  • The split-half analysis is a robust method for assessing the stability of the observed connectivity patterns.
  • The statistical analysis using t-tests with Bonferroni correction ensures rigorous evaluation of the group differences.
  • The figure would be strengthened by including effect size measures to quantify the magnitude of the connectivity differences.
Numeric Data
  • Average Connectivity Strength (Encroaching Salience - Salience, First Half): Z(r)
  • Average Connectivity Strength (Encroaching Salience - Runner-up, First Half): Z(r)
  • Average Connectivity Strength (Encroaching Salience - Salience, Second Half): Z(r)
  • Average Connectivity Strength (Encroaching Salience - Runner-up, Second Half): Z(r)
  • P-value (t-test): 0.001
Figure Extended Data Fig. 8

Extended Data Fig. 8 investigates the cortical representation of the salience network in adults with late-onset depression (LOD). It presents a bar graph comparing the percentage of cortex occupied by the salience network in five individuals with LOD and a control group of 37 healthy individuals. The graph shows that the salience network occupies a significantly larger portion of the cortex in individuals with LOD compared to healthy controls.

First Mention

Text: "A similar effect was observed in adults with late-onset depression (Extended Data Fig. 8)."

Context: This sentence, located in the 'Salience network topology is trait-like' section on page 5, refers to the finding that salience network expansion is also present in adults with late-onset depression, similar to the observation in children who later developed depression.

Relevance: This figure extends the main finding of salience network expansion to a specific subgroup of individuals with depression, those with late-onset depression. It suggests that this atypical network topology is not limited to early-onset depression but may be a more general feature of the disorder, regardless of age of onset.

Critique
Visual Aspects
  • The bar graph is clear and concise, effectively conveying the difference in salience network size between the two groups.
  • The use of error bars provides a visual representation of the variability within each group.
  • The figure would benefit from including the exact values of the means and standard deviations in the caption or on the graph itself.
Analytical Aspects
  • The use of a permutation test ensures robust statistical evaluation of the group difference.
  • The figure would be strengthened by reporting the effect size (e.g., Cohen's d) to quantify the magnitude of the difference.
  • The small sample size of individuals with LOD (n=5) limits the generalizability of the findings and warrants further investigation with larger samples.
Numeric Data
  • Percentage of Cortex Occupied by Salience Network (LOD): %
  • Percentage of Cortex Occupied by Salience Network (HC): %
  • P-value (Permutation Test): 0.009
  • Z-score: 2.9
  • Number of Subjects (LOD): 5
Figure Extended Data Fig. 9 | Dense-sampling of depressive symptoms and functional connectivity in a second individual with depression. a,

Extended Data Fig. 9 focuses on the relationship between depressive symptoms and functional connectivity in a second individual with depression (SIMD-6). Panel (a) presents a heatmap illustrating fluctuations in anhedonia-related symptoms over time, derived from various clinical scales. Panel (b) shows brain images depicting the functional connectivity of salience network voxels in the nucleus accumbens during periods of low and high anhedonia. Panel (c) displays histograms illustrating the stability of the correlation between nucleus accumbens-anterior cingulate functional connectivity and anhedonia, assessed using bootstrap resampling.

First Mention

Text: "An identical analysis in SIMD-6, involving 39 study visits with clinical and fMRI data over 8 months, replicated this effect (Fig. 4c,d and Extended Data Fig. 9a,b)."

Context: This sentence, located in the 'Connectivity state predicts anhedonia' section on page 8, refers to the replication of the finding that functional connectivity between salience network nodes is correlated with changes in anhedonia over time. It specifically mentions the analysis in SIMD-6, which is detailed in Extended Data Fig. 9.

Relevance: This figure provides further support for the link between functional connectivity within the salience network and fluctuations in anhedonia symptoms. It replicates the findings from SIMD-4 in a second individual with depression, strengthening the evidence for the role of this circuit in anhedonia.

Critique
Visual Aspects
  • The heatmap in panel (a) effectively visualizes the temporal dynamics of anhedonia symptoms.
  • The brain images in panel (b) clearly show the differences in functional connectivity patterns during low and high anhedonia.
  • The histograms in panel (c) provide a clear representation of the stability of the correlation.
  • The figure would benefit from clearer labeling of the brain regions in the brain images.
Analytical Aspects
  • The use of dense sampling allows for a detailed examination of the relationship between symptoms and connectivity over time.
  • The bootstrap resampling analysis provides a robust assessment of the stability of the correlation.
  • The figure would be strengthened by quantifying the effect size of the correlation between connectivity and symptoms.
Numeric Data
  • Number of Study Visits (SIMD-6): 39
  • Time Period (SIMD-6): 8 months
  • Correlation (NAc-ACC FC and Anhedonia, SIMD-6): r
  • Number of Bootstrap Repetitions: 1000
  • Average Correlation (Bootstrap Resampling, SIMD-6): r
Figure Extended Data Fig. 10

Extended Data Figure 10 presents a visual representation of the long-term assessment of anhedonia and anxiety-related symptoms in two deeply-sampled individuals with major depression, SIMD-4 and SIMD-6. The figure features two panels, one for each individual, each containing a heatmap that illustrates fluctuations in the severity of 27 clinical symptoms related to anxiety over time. The heatmaps use a color gradient, with blue indicating lower severity and red indicating higher severity. The figure also includes line graphs depicting the fluctuations in anhedonia (blue) and anxiety (red) over time for each individual. The caption mentions that the first principal component (PC1) of the anhedonia and anxiety measures were modestly correlated with one another within each individual over time (Pearson correlation, MDD04: r = 0.41, P < 0.001; MDD06: r = 0.45, P < 0.001).

First Mention

Text: "See Extended Data Fig. 10 for stacked anhedonia and anxiety symptom heatmaps."

Context: This sentence appears in the 'Connectivity state predicts anhedonia' section on page 8, within the context of discussing the dissociability of anhedonia and anxiety symptoms in depression.

Relevance: This figure is relevant to the section's discussion of the relationship between salience network connectivity and specific symptom domains in depression. It provides visual evidence that anhedonia and anxiety symptoms can fluctuate independently over time, supporting the notion that different neural circuits may be involved in these distinct symptom domains.

Critique
Visual Aspects
  • The heatmaps effectively convey the fluctuations in symptom severity over time, with the color gradient providing a clear visual representation of changes in severity.
  • The inclusion of line graphs for anhedonia and anxiety provides a complementary visualization of the overall trends in symptom severity.
  • The figure could be improved by adding specific time points on the x-axes of the heatmaps to allow for more precise alignment of symptom fluctuations with specific dates or events.
Analytical Aspects
  • The figure provides valuable insights into the temporal dynamics of anhedonia and anxiety symptoms in individual patients, highlighting the heterogeneity of symptom expression over time.
  • The modest correlation between the PC1 of anhedonia and anxiety measures suggests that these symptoms can fluctuate independently, supporting the need for investigating their distinct neural correlates.
  • The figure would benefit from a more detailed analysis of the specific clinical symptoms contributing to the observed fluctuations in anxiety, potentially revealing patterns or clusters of symptoms that co-vary over time.
Numeric Data
  • Pearson correlation between PC1 of anhedonia and anxiety measures in SIMD-4: 0.41
  • P-value for Pearson correlation in SIMD-4: 0.001
  • Pearson correlation between PC1 of anhedonia and anxiety measures in SIMD-6: 0.45
  • P-value for Pearson correlation in SIMD-6: 0.001
  • Number of anxiety-related clinical symptoms visualized in the heatmaps: 27

Discussion

Overview

This Discussion section interprets the study's findings on salience network expansion in depression, highlighting its potential as a trait-like marker for depression risk and discussing possible underlying mechanisms. It also addresses the implications of the findings for understanding the interplay between trait and state effects in depression, the potential clinical utility of salience network expansion as a biomarker, and the need for further research to explore causal relationships and specificity to depression. The section concludes by emphasizing the value of precision functional mapping and longitudinal sampling in advancing our understanding of depression and developing personalized treatments.

Key Aspects

Strengths

Suggestions for Improvement

Methods

Overview

This Methods section meticulously outlines the datasets, MRI acquisition parameters, preprocessing steps, and analysis techniques employed in the study investigating salience network topology in individuals with depression. It provides a comprehensive description of the participant groups, including the SIMD dataset of deeply sampled individuals with depression and various healthy control datasets. The section details the multi-echo and single-echo fMRI acquisition protocols, anatomical preprocessing, cortical surface generation, and denoising procedures for both multi-echo and single-echo data. It further explains the surface processing, CIFTI generation, precision mapping of functional brain networks using the InfoMap algorithm, calculation of network size and spatial locations, classification analysis, evaluation of network displacement, assessment of network stability over time, and longitudinal analyses relating connectivity changes to symptom severity.

Key Aspects

Strengths

Suggestions for Improvement

↑ Back to Top