diff --git a/blog/2024-1-2-new-year.md b/blog/2024-1-2-new-year.md new file mode 100644 index 0000000..6a0e92f --- /dev/null +++ b/blog/2024-1-2-new-year.md @@ -0,0 +1,56 @@ +--- +title: New Year Updates +authors: alejandro +tags: [neurosynth] +--- +Hello Neurosynth Users, + +Happy New Year! + +2023 was a very exciting year for Neurosynth, having launched our Compose platform to the public and announced it on social media. In the December we’ve saw **over 500 new user visits**, with **200 users signing up for an account**! 🚀 + +Help us keep this growth going by [sharing our announcement](./blog/announcing-ns-compose) with your colleagues. 🧑‍🔬 + +# 🌟 What’s New 🌟 + +We’ve also continued to introduce new features and improve the user experience. Here’s some highlights: + +### Large-scale association tests + +A key feature that set Neurosynth aside were large-scale association maps (previously known as “reverse inference”). + +Whereas a typical meta-analysis tells you if activity is consistently reported in a target set of studies, an association test tells you if **activation occurs more consistently in this set of studies versus a large and diverse reference sample**. + +That's important, because this allows you to control for base rate differences between regions. Certain regions, such as the insula or lateral PFC for instance, play a very broad role in cognition, and hence are consistently activated for many different tasks and cognitive states. Thus, if you see insula activity in your meta-analysis, you might erroneously conclude that the insula is involved in the cognitive state you're studying. A large-scale association test lets you determine if the activity you observe in a region occurs *more consistently* in your meta-analysis than in other studies, making it possible to make more confident claims that a given region is involved in a particular process, and isn't involved in just about every task. + +Previously association tests were available for the automatically generated maps on neurosynth.org. **Now you can perform large-scale association tests for your custom meta-analyses in Neurosynth Compose.** + +We have created a full primer and tutorial on MKDA Chi-Squared, including an example from a recent meta-analysis on social processing. Check it out! + +import Button from '@mui/material/Button'; + + + +### UX Enhancements ✨ + +Based on your valuable feedback, we've made numerous bug fixes and improvements: + +* **Simplified Curation**: The review import page has been removed, and summary information is now added directly to the tag step. + +* **Searching UI**: We've replaced the dropdown with a selection gallery, making it easier to choose your preferred search method, and we now auto-generate search import names. In addition, resolving duplicates is skipped if none are present. + +* **Improved Editing Workflow**: The editing interface has been improved, streamlining the extraction process. + +* **Various UX Improvements and Fixes**: We fixed many papercuts, especially in the *Extraction* phase. + + +We hope you enjoy these changes. + +Email us any [feedback](mailto:neurosynthorg@gmail.com), or ask a question on [NeuroStars](https://neurostars.org/tag/neurosynth-compose) if you have issues. + + +Cheers, + +The Neurosynth Team 🧠 diff --git a/docs/guide/walkthrough/Studies/Searching.md b/docs/guide/Explore/Searching.md similarity index 100% rename from docs/guide/walkthrough/Studies/Searching.md rename to docs/guide/Explore/Searching.md diff --git a/docs/guide/Explore/index.mdx b/docs/guide/Explore/index.mdx new file mode 100644 index 0000000..09a3fdd --- /dev/null +++ b/docs/guide/Explore/index.mdx @@ -0,0 +1,26 @@ +--- +title: Explore +sidebar_position: 1 +--- + +# Explore + +Here, you can browse and search existing public `Studies`, `StudySets` and `Meta-Analyses` created on the platform. + +## Studies + +The `Studies` page lets you browse and search all studies that exist on the NeuroStore server. This interface is similar to what you'll see when importing studies into your `Project`. However, here it's simply provided for your browsing pleasure. + +For more information on how advanced search functionally, see [Searching Studies](./Explore/Searching) + +## StudySets and Meta-Analyses + +For `StudySets` and `Meta-Analyses`, you can browse and search any user-contributed items, including those from other users. + +Note that although you see all publically available items, you cannot edit somebody else's content. + +:::note +We are currently working on a way to allow users to fork other users' `StudySets` and `Meta-Analyses` to create their own versions. +Stay tuned! +:::: + diff --git a/docs/guide/walkthrough/Project/Curation.md b/docs/guide/Project/Curation.md similarity index 100% rename from docs/guide/walkthrough/Project/Curation.md rename to docs/guide/Project/Curation.md diff --git a/docs/guide/walkthrough/Project/Extraction.md b/docs/guide/Project/Extraction.md similarity index 97% rename from docs/guide/walkthrough/Project/Extraction.md rename to docs/guide/Project/Extraction.md index 71c52cf..bc9aee3 100644 --- a/docs/guide/walkthrough/Project/Extraction.md +++ b/docs/guide/Project/Extraction.md @@ -15,7 +15,7 @@ After the curation phase is complete, the user is redirected to the extraction p Here, the extraction phase starts when a wizard that pops up and guides the user through the process of initializing the extraction phase. On top of creating the initial [**annotation columns**](./Extraction#annotations), this wizard also guides the user through the -process of [**ingestion** ](./Extraction#ingestion) of the curated studies to create a new [**studyset**](../../glossary#studyset). +process of [**ingestion** ](./Extraction#ingestion) of the curated studies to create a new [**studyset**](../glossary#studyset). ## Ingestion diff --git a/docs/guide/walkthrough/Project/Specification.md b/docs/guide/Project/Specification.md similarity index 100% rename from docs/guide/walkthrough/Project/Specification.md rename to docs/guide/Project/Specification.md diff --git a/docs/guide/walkthrough/Project/index.mdx b/docs/guide/Project/index.mdx similarity index 82% rename from docs/guide/walkthrough/Project/index.mdx rename to docs/guide/Project/index.mdx index 2fe8bae..5fed978 100644 --- a/docs/guide/walkthrough/Project/index.mdx +++ b/docs/guide/Project/index.mdx @@ -11,9 +11,8 @@ Within a project you will be able to: 1. **[Curate](./Project/Curation)** studies of interest and select the ones to be included in the meta-analysis 2. **[Extract](./Project/Extraction)** the relevant data such as activation coordinates and other meta-data 3. **[Specify](./Project/Specification)** the algorithm and corrector you would like to use - 4. **Run** the meta-analysis and **View** the results -In each project, you can define a define a single StudySe (i.e. a collection of related studies), and one or more MetaAnalysis specifications. +In each project, you can define a define a single StudySet (i.e. a collection of related studies), and one or more MetaAnalysis specifications. You can open a specific project by logging in, navigating to the diff --git a/docs/guide/Running/index.mdx b/docs/guide/Running/index.mdx new file mode 100644 index 0000000..890524f --- /dev/null +++ b/docs/guide/Running/index.mdx @@ -0,0 +1,85 @@ +--- +title: Running Analyses +sidebar_position: 2 +--- + +# Running Analyses + +You have a several options for running the analysis. In all cases, you will need your unique ``, which you can access for each Meta-Analysis within your Project. + +![Meta-analysis run](/tutorial/ma_run.png) + +Under the hood, analyses are managed by the [nsc-runner](https://github.com/neurostuff/nsc-runner) Python package, and executed by the [NiMARE](https://nimare.readthedocs.io/en/stable/) (Neuroimaging Meta-Analysis Research Environment) Python package. + +## Google Colab + +[![text](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/neurostuff/neurosynth-compose-notebook/blob/main/run_and_explore.ipynb) + +The easiest way to run an analysis is to use the [Google Colab](https://colab.research.google.com/) notebook linked above. + +The provided notebook runs entirely in the cloud, and does not require any local installation of software. +To use simply paste your analysis ID into the first cell (`META_ID`), and using the Toolbar selet (Runtime -> Run All) +or the keyboard shortcut (Ctrl or ⌘ + F9) to run the notebook. + +![Colab notebook](/guide/nsc_colab_notebook.png) + +The notebook will install all required software, run the analysis, and upload the results to Neurosynth Compose. +Once the analysis is complete, you can use the notebook to explore the results using the interative report, download an archive +of the results, or browse the results in the Neurosynth Compose web interface, in the Meta-Analysis section of your Project. + +:::tip +The Colab notebook has limited and varying freely available resources, and may not be able to run large analyses. +If your analysis fails, try running it again, or using one of the other methods below. +::: + +## Docker + +The easiest way to run analyses locally is to use the `nsc-runner` [Docker](https://www.docker.com/) image provided by Neurosynth Compose. + +Docker is a containerization technology that allows you to run software in a consistent environment, regardless of the underlying operating system. + +To run the Docker image, you will need to install Docker on your local machine. +Instructions for installing Docker can be found [here](https://docs.docker.com/get-docker/). + +Once Docker is installed, you can run your analysis using the using the following command: + +``` +docker run -it -v -v /local/dir:/results ghcr.io/neurostuff/nsc-runner:latest +``` + +where `/local/dir` is the path to a local directory where you would like to save the results of your analysis, and `` is the ID of the meta-analysis you would like to run. + +The Docker image will download all required software, run the analysis, and upload the results to Neurovault & Neurosynth Compose. +An HTML report will be saved in the results directory, and the results will be available in the Meta-Analysis section of your Project on Neurosynth Compose. + +### Updating the Docker image + +For every release of `nsc-runner`, we publish a corresponding Docker image. + +You can manually download a specific neuroscout-cli release as follows: + +``` +docker pull ghcr.io/neurostuff/nsc-runner: +``` + +where `` is the version of `nsc-runner` that you want to download. If you omit version, the latest stable version will be downloaded. + +You can see the tags available for download on [GitHub](https://github.com/neurostuff/compose-runner/pkgs/container/nsc-runner) + +## Manually prepared environment using pip + +:::warning +Manually installing `nsc-runner` may be difficult due to complex dependencies in the SciPy stack, or fMRI-specific tooling. Proceed only if you know what you’re doing. +::: + +Use pip to install `nsc-runner` from PyPI: + +``` +pip install nsc-runner +``` + +and then run the analysis using the following command: + +``` +nsc-runner +``` diff --git a/docs/guide/glossary.md b/docs/guide/glossary.md index ae0810a..bc475a1 100644 --- a/docs/guide/glossary.md +++ b/docs/guide/glossary.md @@ -1,6 +1,6 @@ --- title: Glossary -sidebar_position: 1 +sidebar_position: 3 --- # Glossary diff --git a/docs/guide/walkthrough/Studies/index.mdx b/docs/guide/walkthrough/Studies/index.mdx deleted file mode 100644 index c4cc0e4..0000000 --- a/docs/guide/walkthrough/Studies/index.mdx +++ /dev/null @@ -1,6 +0,0 @@ ---- -title: Studies Page -sidebar_position: 0 ---- - -# Studies Page \ No newline at end of file diff --git a/docs/guide/walkthrough/_category_.json b/docs/guide/walkthrough/_category_.json deleted file mode 100644 index eacaa9c..0000000 --- a/docs/guide/walkthrough/_category_.json +++ /dev/null @@ -1,4 +0,0 @@ -{ - "label": "Walkthrough", - "position": 0 -} diff --git a/docs/guide/walkthrough/index.mdx b/docs/guide/walkthrough/index.mdx deleted file mode 100644 index 8d4ac5d..0000000 --- a/docs/guide/walkthrough/index.mdx +++ /dev/null @@ -1,14 +0,0 @@ ---- -title: Walkthrough -sidebar_position: 0 ---- - -import DocCardList from '@theme/DocCardList'; - -# Walkthrough - -This is a comprehensive walkthrough and reference for each page of Neurosynth Compose. - -If you're looking for a quickstart, check out our [Tutorials](/tutorial). - - \ No newline at end of file diff --git a/docs/guide/walkthrough/meta-analyses.md b/docs/guide/walkthrough/meta-analyses.md deleted file mode 100644 index ab3cc24..0000000 --- a/docs/guide/walkthrough/meta-analyses.md +++ /dev/null @@ -1,18 +0,0 @@ ---- -title: Meta-Analyses Page -sidebar_position: 2 ---- - -# a - -## a.a - -## a.b - -## a.b.a - -# b - -## b.b - -# c \ No newline at end of file diff --git a/docs/guide/walkthrough/search.md b/docs/guide/walkthrough/search.md deleted file mode 100644 index 913bba1..0000000 --- a/docs/guide/walkthrough/search.md +++ /dev/null @@ -1,14 +0,0 @@ ---- -title: Searching -sidebar_position: 4 ---- - -# Intro To Searching - -### Searching Studies - -Neurosynth-Compose supports multiple ways to search for studies. - -### Searching Studysets - -### Searching Meta-Analyses \ No newline at end of file diff --git a/docs/guide/walkthrough/studysets.md b/docs/guide/walkthrough/studysets.md deleted file mode 100644 index 83e7701..0000000 --- a/docs/guide/walkthrough/studysets.md +++ /dev/null @@ -1,4 +0,0 @@ ---- -title: Studysets Page -sidebar_position: 1 ---- \ No newline at end of file diff --git a/docs/tutorial/advanced/index.mdx b/docs/tutorial/advanced/index.mdx new file mode 100644 index 0000000..7e15190 --- /dev/null +++ b/docs/tutorial/advanced/index.mdx @@ -0,0 +1,12 @@ +--- +title: Advanced tutorials +sidebar_position: 3 +--- + +import DocCardList from '@theme/DocCardList'; + +# Advanced tutorials + +After you've completed the core Manual and Advanced tutorials, you can continue your learning journey with these advanced tutorials. + + diff --git a/docs/tutorial/advanced/mkda_association.md b/docs/tutorial/advanced/mkda_association.md new file mode 100644 index 0000000..259d1f8 --- /dev/null +++ b/docs/tutorial/advanced/mkda_association.md @@ -0,0 +1,142 @@ +--- +sidebar_label: 'MKDA Chi-Squared Association' +sidebar_position: 3 +--- + +# MKDA Chi-Squared and large-scale association tests +*How to perform large-scale association tests using MKDA Chi-Squared Meta-Analysis, with a Social Processing example* + +import { Card, CardContent, Typography, Button } from '@mui/material'; +import { FaDownload } from 'react-icons/fa' + +## The Reverse Inference Problem + +A common goal of neuroimaging meta-analysis, is to pool a set of studies that invoke common psychological constructs to identify where brain activity is consistently activated. + +Although this is a useful approach, there is a significant inferential challenge-- namely, determining how *specific* the relationship between activity in a given region and the cognitive state invoked by the target task. Ideally, we would like to infer the probability of a mental state given activity in a given region. However, this is exceedingly difficult due to the well-established problem of *reverse inference* (Poldrack, 2011). + +Even if we establish that a given task (e.g. working memory) activates a region 100% of the time (e.g. lateral prefrontal cortex), this observation only establishes that working memory engagement is a sufficient condition for LPFC activity, but *not* that LPFC activity indicates working memory engagement (Poldrack & Yarkoni, 2015). In practice, we know that brain regions are activated by a variety of cognitive processes, and that certain regions of the brain- such as the insula, lateral PFC and medial frontal cingulate cotex (MFCC)- have a high base rate of activation, making it difficult to establish specificity. Using the Neurosynth database (15,000+ studies), we can map the probability of activation of all voxels. Across this large and diverse dataset, certain voxels in MFCC and insula are activate in as many as 20% of studies. + +![Prob-A](/tutorial/prob-A_neurosynth.png) +*Probability of Activity for all Voxels across the Neurosynth Dataset* + +The reverse inference problem is a challenge even for rigorous, high-quality meta-analyses. For example, a recent meta-analysis of RDoC social constructs across 864 fMRI contrasts, [Pintos Lobo et al., (2022)](https://pubmed.ncbi.nlm.nih.gov/36436737/) found converging activation across a variety of regions for "All Social Processing Tasks", including mPFC, ACC, PCC, TPJ, bilateral insula, amygdala, fusiform gyrus, precuneus, and thalamus. However, some of the regions have a high base rate of activation, making it difficult to know how strongly associated their activity is with social processing. + +![Lobos Pinto](/tutorial/pinto_lobos_figa.png) + +*Fig 5a (condensed) from Pintos Lobo et al., (2022). Convergent Activation Patterns Across all social processing tasks (864 contrasts, 1,109 total annotations). * + +Although reverse inference poses a serious challenge, there are certain questions we can ask using large-scale meta-analytic databases that can help. Specifically: **does activity occur *more consistently* for studies that elicit by the mental construct of interest (in this case, social processing) than studies that *do not* elicit that construct** Large-scale meta-analytic datasets can serve as a useful reference, as they consists of tens of thousands of diverse neuroimaging studies automatically sampled from the literature. + +## MKDA Chi-Squared + +We can answer this question using a `Multilevel kernel density (MKDA) analysis - Chi-square` analysis, originally introduced in [Wager et al.,](https://doi.org/10.1093/scan/nsm015). For every voxel, we test if a greater proportion of studies in our meta-analysis activate a given voxel than in a large set of studies that *we did not select* for our inclusion criteria. + +Conceptually, this tests if there's evidence of a *population level* association between the task or psychological construct in our meta-analysis and brain activation (for every voxel). It is equivalent to conducting a chi-squared test of independence for a 2-by-2 table of counts for each voxel, where the binary variables are foci occurrence in the meta-analysis of interest and foci occurrence in the reference set of unselected studies. + +## How to run MKDA Chi-Squared on Neurosynth Compose +### Specification + +Specifying an MKDA Chi-Square meta-analysis in Neurosynth is easy. Simply, select a target set of Analyses to include from your StudySet as you would for any other meta-analysis. + +In Step 3 ("Create Meta-Analysis Specification") of your Project, select *MKDAChi2* as the *algorithm*. + +![MKDA Chi Squared](/tutorial/mkda_chi_squared_algo.png) + +:::note +By default, the `FDRCorrector` is selected, which will perform cluster correction using False Detection Rate with an *alpha* of 0.05. +This is a fast algorithm, however, it is recommended to use `FWECorrector` (family-wise-error) with the `montecarlo` method for more accurate, publication-quality results. +::: + +Next, select the annotation inclusion column you want to use, as before (by default, the "included" column will be used). + +Now, select a reference dataset from the dropdown list below. The Neurosynth dataset represents the latest release of the legacy *Neurosynth* dataset (version 7), released July, 2018. The *Neurostore* dataset represents the latest update of our continuously updating "live" dataset, spanning over 20,000 neuroimaging studies. + +![MKDA Chi Squared Reference](/tutorial/mkda_chi_squared_reference.png) + +Now simply complete the rest of the meta-analysis specification wizard to finish. + +### Executing your analysis + +As usual, you can execute your meta-analysis using Google Colab or on a local computational resource using Docker. + +:::tip +The `MKDAChi2` algorithm takes between ~30s-2minutes to run. However, the `FWECorrector` with 5,000+ montecarlo iterations can take several hours to complete. +We recommend using a workstation or HPC and specifying `--n-cores` at run-time. +::: + +## Interpreting results + +The *MKDA Chi-Squared* Workflow outputs two key maps: **uniformity** and **association** test maps. + +- **Uniformity test map:** z-scores from a one-way ANOVA testing whether the proportion of studies that report activation at a given voxel differs from the rate that would be expected if activations were uniformly distributed throughout gray matter. + +The uniformity test map can be interpreted in roughly the same way as most standard whole-brain fMRI analysis: it displays the degree to which each voxel is consistently activated in studies that use a given term. For instance, for a meta-analysis of "emotion" high z-scores in the amygdala implies that studies that use the word emotion a lot tend to consistently report activation in the amygdala--at least, more consistently than one would expect if activation were uniformly distributed throughout gray matter. + +- **Association test map**: z-scores from a two-way ANOVA testing for the presence of a non-zero association between term use and voxel activation. + +The association test maps tell you whether activation in a region **XXX** occurs more consistently for studies in your meta-analytic sample **m** than for other studies in the reference dataset. In other words, a large positive z-score implies that studies in a meta-analysis are more likely to report **XXX** activation than studies whose abstracts don't include the word 'emotion'. + +Note that association maps *do not* tell you what the probability of a given psychological concept or task is. High Z-scores do not imply that a certain region or voxel is *selective* for a given concept or task. Instead, it just means there is evidence that there is at least a non-zero difference between reference studies, and studies in the meta-analysis. + +:::note +*NiMARE* outputs a variety of maps, including cluster-corrected and uncorrected versions of all maps. + +See the documentation sections on [Outputs of NIMARE](https://nimare.readthedocs.io/en/stable/outputs.html) and [Monte Carlo multiple comparisons](https://nimare.readthedocs.io/en/stable/cbma.html#the-monte-carlo-multiple-comparisons-correction-method) for more details. +::: + +## Example: Pintos Lobo (2022) - All Social Processing Tasks + +To demonstrate, we used Neurosynth-Compose to replicate the [Pintos Lobo et al., (2022)](https://pubmed.ncbi.nlm.nih.gov/36436737/) meta-analysis for All Social Processing Tasks. For this example, we have already created a `Project` and `StudySet` with the coordinates used in this meta-analysis. + +We then specified a `MKDAChi2` Meta-Analysis with `FWECorrector` with the `montecarlo` method with 5,000 iterations. + + + + +### Results + +First, let's look at the FWE cluster corrected **uniformity test** map. + +`z_desc-uniformityMass_level-cluster_corr-FWE_method-montecarlo` +![Uniformity](/tutorial/pinto_lobos_z_desc-uniformityMass_level-cluster_corr-FWE_method-montecarlo.nii.gz.png) + +In this analysis, we replicate the findings of Pinto Lobos (2022), showing consistent activation for social processing across a variety of regions. + +Next, let's look at the FWE cluster corrected **association map**: + +`z_desc-associationMass_level-cluster_corr-FWE_method-montecarlo` +![Association](/tutorial/pinto_lobos_z_desc-associationMass_level-cluster_corr-FWE_method-montecarlo.nii.gz.png) + +As before, regions which have been previously implicated with social processing, such as the tempo-parietal junction (TPJ), and dorso-medial and ventro-medial PFC are present, meaning that activity in these social processing studies report activity in these regions with greater frequency than other studies in the Neurosynth database. + +However, certain regions which we know to have low specificity, such as the insula, medial frontal cingulate cortex (MFCC) and parts of dorso-lateral PFC, are absent, meaning that there is *no evidence* that social processing tasks report activity in these regions *more frequently* than other studies in the database. + +This example demonstrates how `MKDA Chi-Squared` association analysis can help determine the specificity activity and tasks in a meta-analysis, even for high-quality manual meta-analyses. + + +## Footnotes & Limitations + +**What happened to the "forward inference" and "reverse inference" maps?** + +We renamed the pre-generated forward and reverse inference maps; they're now referred to as the "uniformity test" and "association test" maps that we discuss here. + +Although the method we used hasn't changed (`MKDA Chi-Squared`), the latter names more accurately capture what these maps actually mean. It was a mistake on our part to have used the forward and reverse inference labels; those labels should properly be reserved for posterior probability maps generated via a Bayesian estimation analysis, rather than for z-scores resulting from a frequentist inferential test of association. Probability maps are more difficult to interpret and use correctly, as they depend on the *prior* assumed by the researcher. Since setting an appropriate prior is highly non-trivial, these maps are disabled by default. + +** Using MKDA Chi Squared on manual meta-analyses ** + +In this tutorial, we applied `MKDA Chi-Squared` to a manual meta-analysis. However, this is not a perfect comparison, as there are differences between the reference sample (Neurosynth), the high-quality manual annotations given as input. Studies in large-scale meta-analytic databases are automatically populated, meaning there are potential sampling biases. Most notably, studies in Neurosynth include all reported coordinates, not only "target" analyses/contrasts. Thus, it is possible that low-level task > no task contrasts are over-represented in this reference sample. + +## References & Further Reading + + +If you want to understand the nuances of what inferences you can and cannot make using these maps, we recommend reading Tal Yarkoni's blog posts on how these maps do not provide evidence that the dACC is select for pain: [Post 1](https://www.talyarkoni.org/blog/2015/12/05/no-the-dorsal-anterior-cingulate-is-not-selective-for-pain-comment-on-lieberman-and-eisenberger-2015/), [Post 2](https://www.talyarkoni.org/blog/2015/12/14/still-not-selective-comment-on-comment-on-comment-on-lieberman-eisenberger-2015/), as well as a commentary by [Tor Wager et al., 2016](https://www.pnas.org/doi/10.1073/pnas.1600282113) + +Poldrack RA. Inferring mental states from neuroimaging data: from reverse inference to large-scale decoding. Neuron. 2011 Dec 8;72(5):692-7. doi: 10.1016/j.neuron.2011.11.001. PMID: 22153367; PMCID: PMC3240863. + +Poldrack RA, Yarkoni T. From Brain Maps to Cognitive Ontologies: Informatics and the Search for Mental Structure. Annu Rev Psychol. 2016;67:587-612. doi: 10.1146/annurev-psych-122414-033729. Epub 2015 Sep 21. PMID: 26393866; PMCID: PMC4701616. + + + diff --git a/docs/tutorial/automated.md b/docs/tutorial/automated.md index 86fbfd5..0520b77 100644 --- a/docs/tutorial/automated.md +++ b/docs/tutorial/automated.md @@ -1,9 +1,10 @@ --- sidebar_label: 'Automated Meta-Analysis' -sidebar_position: 3 +sidebar_position: 2 --- # Automated Meta-Analysis +*How to create a fully automated meta-analysis* :::caution This tutorial, and the functionality for automated meta-analysis, is under construction. diff --git a/docs/tutorial/index.mdx b/docs/tutorial/index.mdx index 74bedb1..3410a36 100644 --- a/docs/tutorial/index.mdx +++ b/docs/tutorial/index.mdx @@ -5,11 +5,15 @@ import { Grid, Card, CardContent, Typography, Button } from '@mui/material'; import { Link } from 'react-router-dom'; +import DocCardList from '@theme/DocCardList'; -Neurosynth Compose supports a range of different workflows, from exploratory large-scale automated analyses to highly curated and rigorous manual analyses. +### Quickstart -The choice of workflow depends on the research question and the available resources. Note that you can choose to blend approaches from these two -ends of the spectrum, depending on your needs. Learn more about the different workflows by following the tutorials below. +Neurosynth Compose supports a range of workflows, +from exploratory large-scale automated analyses to highly rigorous manual analyses. + +The choice of workflow depends on your research question and resources available for manual curation. +We reccomend starting with the **manual meta-analysis** tutorial if you are new.
@@ -49,3 +53,8 @@ ends of the spectrum, depending on your needs. Learn more about the different wo
+ + +### Advanced tutorials +After you've completed the core tutorials above, you can continue your learning journey with [advanced tutorials](./tutorial/advanced). + diff --git a/docs/tutorial/manual.md b/docs/tutorial/manual.md index 2b22877..d593e0d 100644 --- a/docs/tutorial/manual.md +++ b/docs/tutorial/manual.md @@ -1,9 +1,10 @@ --- sidebar_label: 'Manual Meta-Analysis' -sidebar_position: 2 +sidebar_position: 1 --- # Manual Meta-Analysis +*How to create a custom, manual meta-analysis.* import { Card, CardContent, Typography, Button } from '@mui/material'; import { FaDownload } from 'react-icons/fa' diff --git a/docusaurus.config.js b/docusaurus.config.js index 10989e3..0b8b63d 100644 --- a/docusaurus.config.js +++ b/docusaurus.config.js @@ -77,8 +77,8 @@ const config = { }, {to: '/blog', label: 'Blog', position: 'left'}, { - href: 'https://github.com/neurostuff/compose-docs', - label: 'GitHub', + href: 'https://compose.neurosynth.org', + label: 'Compose Home', position: 'right', }, ], diff --git a/static/guide/nsc_colab_notebook.png b/static/guide/nsc_colab_notebook.png new file mode 100644 index 0000000..fadd1d8 Binary files /dev/null and b/static/guide/nsc_colab_notebook.png differ diff --git a/static/img/compose_ecosystem_paths.png b/static/img/compose_ecosystem_paths.png new file mode 100644 index 0000000..86d2922 Binary files /dev/null and b/static/img/compose_ecosystem_paths.png differ diff --git a/static/tutorial/mkda_chi_squared_algo.png b/static/tutorial/mkda_chi_squared_algo.png new file mode 100644 index 0000000..4f5c502 Binary files /dev/null and b/static/tutorial/mkda_chi_squared_algo.png differ diff --git a/static/tutorial/mkda_chi_squared_reference.png b/static/tutorial/mkda_chi_squared_reference.png new file mode 100644 index 0000000..489e98b Binary files /dev/null and b/static/tutorial/mkda_chi_squared_reference.png differ diff --git a/static/tutorial/neurosynth_paper_fig2.jpg b/static/tutorial/neurosynth_paper_fig2.jpg new file mode 100644 index 0000000..44fe571 Binary files /dev/null and b/static/tutorial/neurosynth_paper_fig2.jpg differ diff --git a/static/tutorial/pinto_lobos_figa.png b/static/tutorial/pinto_lobos_figa.png new file mode 100644 index 0000000..be91921 Binary files /dev/null and b/static/tutorial/pinto_lobos_figa.png differ diff --git a/static/tutorial/pinto_lobos_z_desc-associationMass_level-cluster_corr-FWE_method-montecarlo.nii.gz.png b/static/tutorial/pinto_lobos_z_desc-associationMass_level-cluster_corr-FWE_method-montecarlo.nii.gz.png new file mode 100644 index 0000000..4416215 Binary files /dev/null and b/static/tutorial/pinto_lobos_z_desc-associationMass_level-cluster_corr-FWE_method-montecarlo.nii.gz.png differ diff --git a/static/tutorial/pinto_lobos_z_desc-uniformityMass_level-cluster_corr-FWE_method-montecarlo.nii.gz.png b/static/tutorial/pinto_lobos_z_desc-uniformityMass_level-cluster_corr-FWE_method-montecarlo.nii.gz.png new file mode 100644 index 0000000..677a751 Binary files /dev/null and b/static/tutorial/pinto_lobos_z_desc-uniformityMass_level-cluster_corr-FWE_method-montecarlo.nii.gz.png differ diff --git a/static/tutorial/prob-A_neurosynth.png b/static/tutorial/prob-A_neurosynth.png new file mode 100644 index 0000000..c4592c0 Binary files /dev/null and b/static/tutorial/prob-A_neurosynth.png differ