Summary Statistics by Tissue, Assay, and Platform

Group- and timepoint-level summary statistics computed across multiple tissues, molecular assays, and analytical platforms. Each dataset contains means and standard deviations calculated within randomization groups and timepoints. These objects are intended for descriptive and exploratory use and are not differential analysis results. Available for all omes/tissue/platforms except methylcap, which is processed separately.

For any datasets with missing values, the values reflect means/sds/n with missing values excluded.

Usage

## Adipose
ADIPOSE_EPIGEN_METHYLCAP_SEQ_SUM_STATS
ADIPOSE_METAB_SUM_STATS

ADIPOSE_PROT_PH_SUM_STATS
ADIPOSE_PROT_PR_SUM_STATS
ADIPOSE_TRANSCRIPT_RNA_SEQ_SUM_STATS

## Blood
BLOOD_EPIGEN_METHYLCAP_SEQ_SUM_STATS
BLOOD_METAB_SUM_STATS

BLOOD_PROT_OL_SUM_STATS
BLOOD_METAB_T_CLINICAL_SUM_STATS
BLOOD_PROT_CLINICAL_SUM_STATS
BLOOD_TRANSCRIPT_RNA_SEQ_SUM_STATS

## Muscle
MUSCLE_EPIGEN_METHYLCAP_SEQ_SUM_STATS
MUSCLE_METAB_SUM_STATS

MUSCLE_PROT_PH_SUM_STATS
MUSCLE_PROT_PR_SUM_STATS
MUSCLE_TRANSCRIPT_RNA_SEQ_SUM_STATS
MUSCLE_EPIGEN_ATAC_SEQ_SUM_STATS

Format

Each object is a data.frame with one row per feature per randomization group and timepoint, carrying the columns below in this order. They are named and ordered to agree with the *_DA objects: what the rows are, then what identifies a row, then the statistics.

tissue

adipose, blood or muscle.

assay

The ome. Every metabolomics platform is stacked under "metab" and named in platform, exactly as the *_DA objects do it; every other ome names itself here.

platform

The metabolomics platform, for example "metab-u-rppos" or "metab-t-clinical". Present on metabolomics objects only, which is also how the *_DA objects carry it. Objects for any other ome do not have this column.

randomGroupCode

Randomization group: ADUControl, ADUEndur, ADUResist.

Timepoint

Timepoint within the acute bout.

feature_id

Feature identifier, in the same namespace as the differential analysis for that tissue and ome.

Count

Number of samples summarised, after excluding missing values.

Mean

Mean across those samples.

SD

Standard deviation across those samples, NA when Count is 1.

Before v2.0.1 the platform was written into assay, there was no platform column, and the identifying columns came first with tissue and assay last. There was also one object per metabolomics platform rather than one per tissue.

Details

Datasets are stratified by tissue (e.g., adipose, blood, muscle) and assay (metabolomics, proteomics, transcriptomics, epigenomics) — one object per tissue and assay, named the way the *_DA objects are named.

Every metabolomics object reads assay = "metab" and names its platform in a platform column (e.g. "metab-u-rppos", "metab-t-tca"), which is how the *_DA objects are labelled.

The research platforms are not one object each. They are stacked into a single {TISSUE}_METAB_SUM_STATS per tissue, which is how {TISSUE}_METAB_DA is keyed, so the two tiers nest the same way. BLOOD_METAB_T_CLINICAL_SUM_STATS is clinical chemistry: it carries the same labelling but stays its own object, on this tier and the differential-analysis tier alike, because it shares five analytes with the research platforms (Cortisol, Glucose, Glycerol, KET, NEFA) that would otherwise sit in one object twice. platform is what tells the two apart — assay does not.