Replication Data for Multifactorial Depression Prediction in 217 Individuals

Replication Data for: Multifactorial Prediction of Depression Diagnosis and Symptom Dimensions is a dataset accompanying a submitted manuscript from the Texas Data Repository. It contains data from 217 individuals classified into currently depressed, psychiatric control, and no psychopathology groups. The data was used to build cross-validated models predicting diagnosis and symptom severity, explaining up to 65.7% of variance in symptom severity.

Use Cases

• Training classification models to distinguish depression from other psychiatric conditions based on multimodal predictors.
• Predicting depression symptom severity scores using sociodemographic, behavioral, and physiological data.
• Analyzing the relative importance of predictors like psychosocial functioning and rumination across different diagnostic models.
• Benchmarking machine learning models for psychiatric classification against a dataset with a psychiatric control group.

Strengths

• Includes a psychiatric control group to help isolate factors specific to Major Depressive Disorder.
• Models built from the data explained up to 65.7% of variance in depression symptom severity.
• Integrates predictors from seven distinct domains, including polygenic scores, EEG, and actigraphy.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is limited to 217 individuals, which may constrain model complexity or generalizability.
• Description metadata is limited; actual data quality and file formats require manual inspection after download.

Provenance

Source

Texas Data Repository Harvested Dataverse, author Mary McNamara.

Collection Method

Data collected for a research study, likely involving participant assessments and measurements.

Time Range

null

Freshness

Last updated 2024-03-18 06:28:10; freshness should be verified.

Geography

null