• Description
• Parameters
  • Working Directory
  • About Data
    • Number of Replicates
  • Smoothing
    • Gaussian Process Variance Parameter
    • Gaussian Process Scale Parameter
  • Gene Set Analysis (GSA)
    • Biological Function Annotation Files
    • Enrichment Analysis FDR Cutoff
    • Minimum % of Genes to Consider a Pathway
    • Minimum Number of Genes For Hypothesis Testing
  • Select Data
    • Gene Name Column
    • Expression Series 1
    • Data Input Form 1
    • Expression Series 2
    • Data Input Form 2
    • mRNA Level Inference
    • mRNA Expression Series
    • Data Input Form
  • MCMC Parameters
    • MCMC Burn-In
    • MCMC Thinning
    • MCMC Samples
• Output
  • General Output
  • GSA Output (if GSA had been checked)

PECA Core implements the core functionality for analyzing a two-level time series data set (e.g. paired protein and mRNA concentration data). It identifies significantly regulated genes at each time point with probability scores of significant change points.

Note that PECA Core does not deconvolute the contributions of changes in synthesis or degradation.

Specifies the directory where input files, output matrices and plots produced by PECA will be saved. It can be specified manually by typing in the path or the folder can be provided by using the "Select" button.

Friendly Reminder: DO NOT SELECT DESKTOP as the tool produces MANY files

Specification of the number of replicate experiments in the datasets for expression series 1 and 2 (default: 1)

If checked, Gaussian Process (GP) smoothing will be applied to the datasets (default: unchecked).

Determines the variation of values from the mean (default 2.0). A small value will result in the function values changing quickly.

Scaling factor that determines the smoothness of the curve (default 1.0). A small value will result in a function that stays close to the mean value.

If checked, a time-dependent functional enrichment analysis will be performed on the output matrix of PECA, specifically on the change point score based on the regulation rate ratio for PECA-core and N. The result will be displayed as an additional output matrix (default unchecked). The resulting matrix reports the biological functions whose members are up or down-regulated at specific time points.

Specifies the file path of the function annotation file that should be used for the time-dependent functional enrichment analysis.

File Format:

1. First column named as ‘Pathwayid’, specifying the pathway IDs, e.g. from Gene Ontology and Consensus Pathway DataBase
2. Second column named as the same name as gene name columns provided in the parameters, specifying the genes involved
3. Third column named as 'pathway', specifying the pathways involved

Sample file with annotations for human genes

Defines the FDR cutoff for which enrichment analysis should use when analyzing biological functions at specific time points (default 0.05, i.e. 5%). The value of this parameter should lie between 0 and 1 (e.g., 0.05, 0.1, 0.2).

Specifies the minimum percentage of genes needed in the experimental data for a pathway to be analyzed (default 0). For instance, if 20% is specified, then at least 20 genes need to be present in the experimental data for a pathway of 100 genes. The value of this parameter should lie between 0 and 100.

Specifies the minimum number of significant genes (within the FDR cutoff) from the experimental data for a particular pathway to be reported (default 1). Anything below this number will be assigned a p-value of 1. The value of this parameter should be a positive integer.

The selected text column will be used as the gene ID identifiers in PECA analysis (default: first text column).

The selected expression/numerical columns that should be used as expression series 1 (typically mRNA concentration data), which comes before expression series 2 (typically protein concentration data) where expression series 1 represents degradation and 2 represents synthesis (default: first half of expression/numeric columns).

RNA ‘expression’ data can be normalized read counts, e.g. FPKM, RPKM, or TPM, from an RNA seq experiment, or signal intensities from a single channel microarray experiment, or actual concentration (if known).

The columns should be ordered by timepoints and then by replicate

Order:

• time point 1 replicate 1
• …
• time point N replicate 1
• time point 1 replicate 2
• …
• time point N replicate 2

Specification for the data input form of Expression Series 1, i.e. what data transformation has been applied already (default: Raw).

• Raw: unprocessed, untransformed data.

• ln: log_e transformed data.

• log_2: log₂ transformed data.

• log_10: log₁₀ transformed data.

• log_custom: log_X transformed data, where X is a specified positive real value

The selected expression/numerical columns that should be used as expression series 2 (typically protein concentration data) which comes after expression series 1 (default: second half of expression/numerical columns).

Protein ‘expression’ values can be protein intensities from mass spectrometry experiments (e.g. Label Free Quantification, LFQ) or spectral counts or actual concentrations (if known).

Same order as Expression Series 1. The number of columns should also match expression series 1.

Specification for the data input form of Expression Series 2 (default: Raw).

Same as Data Input Form 1

If checked, mRNA level inference will be performed, i.e. Expression Series 1 is assumed to be DNA concentrations with 1 as values and Expression Series 2 is set as the given mRNA Expression Series (default: unchecked). If not, Expression Series 1 and 2 are both set by the user as explained above.

If mRNA Level Inference is checked, this is the selected expression/numerical columns that should be used as mRNA expression data.

Specification for the data input form of mRNA Expression Series (default: Raw).

Same as Data Input Form 1

PECA model parameters are estimated using a sampling-based algorithm called MCMC (Markov chain Monte Carlo), which requires the parameters below. All values should be positive integers.

Defines the iterations to be thrown away at the beginning of MCMC run, i.e. the burn-in period (default: 1000).

Defines the interval in which iterations of MCMC are recorded (default: 10).

Defines the total of number of post-burn-in samples to be recorded from MCMC (default: 1000).

The text column is the gene name column provided by Gene Name Column.

The main/expression columns are the log_e transformed Expression Series 1 and Expression Series 2 data sets.

The other numeric columns contain : RY, signedCPSX, FDRX, where X indexes time point (i.e. X=1 refers to the second time point) and Y indexes time point interval starting from 0 (i.e. Y=0 refers to the interval between the first and second time points).

• RY is the rate ratio for the time interval preceding the specified time point (e.g. if Y = 1, then the interval is between time point indices 1 and 2)

• signedCPSX is the change point score with signs indicating up/down regulation.
  Positive sign describes upregulation; negative sign down regulation.

• FDRX is the False Discovery Rate

NOTE: the rate ratio itself does NOT inform on the significance or direction of the change. The DIFFERENCE between consecutive rate ratios (adjacent time intervals) describes the direction of change, and the FDR the significance.

The text column contains: the gene name column provided from Gene Name Column, GO_name, GO_id

• GO_name is the name of the Gene Ontology

• GO_id is the ID of the Gene Ontology

The numeric columns contain: MaxSig(Up), MaxSig(Down), Max(Both), GO_size, GO_size_background, Up(X), Down(X), Sig(X), where X indexes time points corresponding to signedCPSX

• MaxSig(Up) is the maximum value of -log₁₀(Up(X)) for all X

• MaxSig(Down) is the maximum value of -log₁₀(Down(X)) for all X

• Max(Both) is the maximum value of -log₁₀(Sig(X)) for all X

• GO_size is the number of genes in the pathway

• GO_size_background is the number of genes in the pathway that appears in the experimental data

• Up(X) is the p-value calculated from the number of up-regulated genes

• Down(X) is the p-value calculated from the number of down-regulated genes

• Sig(X) is the p-value calculated from the number of up and down-regulated genes