The search page provides multiple ways to find proteins:

If the search yields multiple results, you can click “View Details” for the target protein in the table to examine it.

After searching, you'll see a results table with the following information:

Click "View Details" to access comprehensive protein information:

Protein Information Section:

• Accession: UniProt identifier
• Protein Names: Full descriptive name
• Gene Symbol: Standard gene name
• Organism: Species information
• Length: Protein size in amino acids
• Related PMIDs: Links to relevant publications

Tissue/Cell Line Expression:

Visual representation of palmitoylation across different tissues:

• Bubble Heatmap: Show the quantity and proportion of palmitoylated proteins found in all the tissues and cell lines in the database.

• Bar Chart: This stacked bar chart displays palmitoylation detection across different research studies (PMID). Coloured segments represent positive samples, while grey segments indicate negative samples from each publication.
• TSI Index: Tissue Specificity Index (higher = more specific) TSI = 0: Ubiquitous expression TSI = 1: Highly tissue-specific"

Protein Sequence with Site Annotation:

The sequence display uses color-coded highlighting:

Palmitoylation Sites Details:

Comprehensive table of all known palmitoylation sites:

PhyloP Conservation Scores:

Measures evolutionary conservation at individual bases using phylogenetic p-values:

• Positive scores: Indicate conservation (higher = more conserved)
• Negative scores: Indicate accelerated evolution (lower = faster evolution)
• Near zero: Neutral evolution
• Color coding: Blue shades for conservation, pink shades for acceleration

PhastCons Conservation Scores:

Identifies conserved elements using hidden Markov models:

• Range: 0 to 1
• Close to 1: Highly conserved, likely functional
• Close to 0: Fast evolving, less constrained
• Color coding: Blue shades for high conservation, pink shades for low conservation

Visualization Features:

• Three-base grouping: Each codon position shows scores for 1st, 2nd, and 3rd bases
• Interactive tooltips: Hover to see detailed scores and position information
• Position labels: Colored badges indicate site types (experimental, predicted, etc.)
• Scrollable view: Horizontal scrolling for proteins with many modification sites
• Human (hg38/GRCh38) Dataset:The “100 Vertebrates Alignment and Conservation (100-way)” dataset from the UCSC Genome Browser. This file contains whole-genome sequences from 100 vertebrate species aligned against the human hg38 reference genome.
• Mouse (mm39/GRCm39) Dataset:The “35 Vertebrates Alignment and Conservation (35-way)” dataset from the UCSC Genome Browser. This file contains whole-genome sequences from 35 vertebrate species aligned against the mouse mm39 reference genome.

Example Conservation Charts:

TCGA Mutation Information:

Cancer-related cysteine mutations that may create new palmitoylation sites:

• Mutation position and amino acid change
• Frequency in cancer samples
• Mutation type and functional impact
• Associated cancer types

Browse palmitoylated proteins by data source (top 100 from experimental studies displayed; all from database sources shown):

Top 100 Proteins Browse by species:

All browse views provide consistent protein information:

Species Selection: Choose between Human or Mouse data

Analysis Type:

• Group A vs Group B: Compare any two groups of datasets
• Cancer vs Normal: Specifically compare cancer vs normal tissues (Human only)

Protein Input: Enter UniProt accessions or gene symbols (e.g., P01116, KRAS, TP53)

Available Datasets:

• Group A vs Group B: Compare any two groups of datasets
• Cancer vs Normal: Specifically compare cancer vs normal tissues (Human only)

The system automatically validates input proteins and provides feedback:

• Not Found - Protein or gene not in database
• Species Mismatch - Protein or gene exists but in different species
• No Expression Data - Protein or gene is not expressed in either of the selected groups.

• Continue : Ignore these errors and proceed with analyzing the correct protein or gene.
• Cancel and Modify Input : Return to the previous step to make adjustments.

Click "Run Differential Expression Analysis" to start the analysis process.

The results page provides:

• Summary Statistics: Overview of analyzed proteins and significant findings

• Results Table: Detailed comparison for each protein including:
  • Protein accession and gene symbol
  • Expression ratios for both groups
  • Odds ratio and p-value
  • Significance indicators (*p<0.05, **p<0.01, ***p<0.001)
• Visualization: Interactive bar chart comparing expression ratios:
  • Click "Customize"
  • Customizable colors and selectable visual proteins
  • Click “Apply Changes” to visualize.
  • Visual images can be downloaded and saved by clicking “Download jpg” or “Download SVG”.

Enter Query Protein: Input a Protein ID (e.g., P12345) or Gene Name (e.g., TP53)

Select Analysis Method:

• Fisher's Exact Test: Provides OR, P-value, FDR, and relationship type
• Jaccard Index: Provides Jaccard coefficient and relationship type

Choose Species: Human or Mouse

Select Tissue/Sample Type:

• Mouse: All, Liver, Brain
• Human: All, Tumor

Click "Search Network" to execute the analysis and generate network visualization.

Table Results

• Top 2500 significant interactions displayed
• 25 results per page with pagination
• Columns include protein identifiers, association metrics, and relationship types
• Sortable by association strength (OR or Jaccard)

Network Visualization

• Drag nodes to rearrange network layout
• Hover over nodes for detailed information
• Scroll to zoom in/out of the network
• Download network as SVG for publications

Enter Two Proteins: Input Protein IDs (e.g., P12345) or Gene Names (e.g., TP53) for both proteins

Select Analysis Method:

• Fisher's Exact Test: Provides OR (Odds Ratio), P-value, FDR, and relationship type
• Jaccard Index: Provides Jaccard coefficient and relationship type

Choose Species: Human or Mouse

Select Tissue/Sample Type:

• Mouse: All, Liver, Brain
• Human: All, Tumor

Click "Analyze Relationship" to execute the analysis and generate statistical results with visualization.

Statistical Results Table

• Protein Accession & Gene Symbol: Identifiers for both proteins
• Association Metrics:
  • OR (Odds Ratio): Strength of association (>1 = positive, <1 = negative)
  • P-value: Statistical significance of the relationship
  • FDR: False Discovery Rate corrected P-value
  • Jaccard: Similarity coefficient (0-1, higher = more similar)
• Relation Type:
  • Co-occurrence (C): Proteins tend to appear together in samples
  • Mutual Exclusion (M): Proteins rarely appear together in samples

Heatmap Visualization

• Binary Representation: White (0) = protein absent, Blue (1) = protein present
• Sample Organization: Samples sorted alphabetically for easy comparison
• Tissue Color Coding: X-axis labels colored by tissue/cell type for quick identification
• Protein Labels: Display protein ID with gene symbol in parentheses when available

Statistical Methods

FDR Correction

• Method: Benjamini-Hochberg FDR correction
• Output: adjust P (corrected P-values)
• Significance threshold: adjust P (Logit) < 0.05
• Display: Significant adjust P (Logit) shown in green

Sample Count Columns

Statistical Columns

Significance Interpretation

Interactive Bar Chart

• Access: Click "Chart" button in any result row
• Maroon Bars: Palmitoylated protein counts
• Blue Bars: Non-palmitoylated protein counts
• Comparison: Mutated vs wild-type samples side-by-side
• Data Labels: Exact count values displayed on bars
• Stacked Display: Clear visualization of palmitoylation status distribution

Statistical Panel

• Complete 2×2 contingency table with sample counts
• Both original P-values and FDR-corrected adjust P values
• Clear significance interpretation with color coding
• Regression coefficient with direction indication
• FDR correction status indicator

Species Selection: Choose between Human or Mouse data

Protein Input: Enter UniProt accessions or gene symbols (maximum 100 proteins)

Data Sources: Select from three categories:

• Experimental: Experimentally verified sites from literature
• Database: Curated sites from SwissPalm, CysModDB, dbPTM, PTMD
• Prediction: Computationally predicted sites (High/Medium/Low confidence)

Window Size: Number of amino acids to extract around each palmitoylation site (1-20, default: 6)

Analysis Method:

• Frequency Analysis: Basic amino acid frequency around sites
• Information Content: More sophisticated motif discovery

After configuration, click "Check Proteins & Continue" to validate input proteins.

The validation page shows:

• Valid Proteins - Ready for analysis with available sites
• Proteins with Issues - Missing data or species mismatches
• Detailed breakdown of available data sources for each protein

• This table provides basic information on palmitylated proteins suitable for further analysis
• Select the palmitylated proteins that you wish to analyse in order to proceed to the next step
• To view more detailed information about a specific palmitoylated protein, click 'Search' to navigate to the protein details page.

• This table provides basic information on palmitylated proteins that could not be analysed
• Reasons why proteins in the table cannot proceed to the next step of analysis
• Reason for error: The protein lacks a palmitoylation site; no related protein exists in the database.

On the confirmation page, review parameters and click "Run Analysis" to start motif discovery.

The results page provides:

• Analysis Summary: Overview of proteins analyzed and motifs found
• Protein Details: List of analyzed proteins with site counts and positions

• Motif Logos: Sequence logos for discovered motifs, sorted by E-value
• Consensus Sequences: Representative sequences for each motif

Understanding E-values: Lower E-values indicate more significant motifs (E-value < 0.05 is typically considered significant)

• Base URL: https://palmlab.intelligent-oncology.com/api
• HTTP Method: GET for all endpoints
• Response Formats: JSON (default), CSV, PNG (for visualization endpoints)
• Features: Pagination, Format selection, Image generation, Batch processing

All endpoints return consistent JSON structure when format=json:

{
  "success": true,
  "message": "Analysis completed successfully",
  "query": {
    "proteins": ["P01112", "P07900"],
    "species": "human",
    "format": "json",
    "timestamp": "2024-01-15T10:30:00Z"
  },
  "results": [
    {
      "protein_accession": "P01112",
      "gene_symbol": "HRAS",
      "cancer_ratio": 0.75,
      "normal_ratio": 0.25,
      "odds_ratio": 9.0,
      "p_value": 0.001,
      "significant": true
    }
  ],
  "summary": {
    "total_found": 1,
    "significant_count": 1,
    "note": "Showing top 50 results"
  }
}

{
  "success": false,
  "message": "Protein not found: XYZ123",
  "error": "Invalid protein ID",
  "status": 400,
  "timestamp": "2024-01-15T10:30:00Z"
}