---

Interactive Morphometric Profiling Pipeline for Scleractinian Larvae (Stylophora pistillata)

This folder hosts an automated, interactive Fiji/ImageJ macro pipeline engineered for quantitative structural analysis of brooded cnidarian planulae. The protocol captures localized tissue dimensions (ectoderm thickness) alongside global cross-sectional constants (total envelope and internal cavity areas) to evaluate maternal phenotypic investment across divergent intra-colonial larval morphs.

---

1. Experimental Design & Biological Rationale

Brooding corals like Stylophora pistillata release well-differentiated planulae containing complex tissue layers and maternally translocated dinoflagellate symbionts. To evaluate if parent colonies execute a structural parental bet-hedging strategy—releasing High-Fluorescence (HF) and Non-Fluorescent (NF) larvae equipped with different survival capacities—this computational workflow turns qualitative histology cuts into precise, peer-review-ready dimensional arrays.

Target Biomarkers Scraped:

• Ectoderm Stratum Thickness: Measures physical cellular boundaries at 5 randomized positions across the stratified ciliated periphery.
• Total Larval Area: Traverses the entire external multi-point tissue envelope boundary to establish global cross-sectional area constants.
• Gastrovascular Cavity Lumen Area: Traces the absolute internal free boundary layer where gastrodermal cells face the open central cavity.

Analytical Logic Matrix:

By acquiring these explicit structural variables, the dataset can be expanded downstream within stats pipelines (such as Pandas, R, or JMP) to isolate parameters that cannot be measured directly due to micro-scale technical artifacts:

Derived Morphometric Metric	Operational Formula	Analytical Utility
Living Tissue Surface Mass	Total Area - Cavity Area	Isolates true cellular biomass distribution, removing any variation introduced by larvae expanding or deflating their water volume during fixation.
Volumetric Lumen Expansion Index	Cavity Area / Total Area	Monitors changes in hydrostatic expansion capacities and fluid transport spaces under stress.
Ectodermal Coefficient of Variation (CV)	Standard Deviation(L1…L5) / Ectoderm Mean	Evaluates structural consistency. A low CV provides statistical proof that section cuts are true, undistorted transverse profiles.

---

2. Technical Engineering Choices & Iterative Optimizations

This code underwent several critical iterations to adapt open-source Fiji architectures to proprietary Olympus bioimaging scans:

• The Suffix & Case-Sensitivity Trap: Raw confocal or widefield systems output compound file extensions with fluctuating casing rules (.oir vs .OIR). Standard string match loops skip non-lowercase extensions silently. This was resolved by implementing compound case checking arrays.
• Resolution Pyramid Bypass: Fiji’s Bio-Formats plugin attempts to save workstation RAM by defaulting to downsampled intermediate layout maps or navigation thumbnails (e.g., loading low-res arrays at 384x355 instead of full-scale multi-megapixel captures). The macro forces native metadata calibration lookups to lock in deep resolution matrices with explicit micrometer scales.
• Physical Dimension Bounds: Direct vector plotting across the acellular connective mesoglea introduces extreme human bias at standard magnifications due to its narrow width (~1–2 µm). This layer’s thickness has been omitted from manual tracing to prevent experimental noise and is monitored implicitly through global polygon area subtraction.
• Node Conflict Resolution: Sequential execution of coordinate collection tools (polygon) on the same frame window causes ImageJ to interpret new mouse clicks as an edit to the previous selection’s anchor points. The macro maps down clear deselection sweeps (Select None and deselect) between passes, allowing operators to drop inside nodes without distorting the outer larva perimeter line.
• Cloud Sync Violations Locked: Direct live updates on cloud networks (like Google Shared Drives) invoke automated background data indexing. Long-term writing processes prompt file collisions, throwing fatal java.io.FileNotFoundException path lock failures. Shifting the storage mechanism to Streaming Disk Appending maps transactions safely on local drives while keeping an internal text index parser (indexOf) to enable Instant Auto-Resume capacities if a multi-image session is interrupted.

---

3. Production Fiji Macro (Planulae_Morphometrics.ijm)

Paste this stable script configuration into Fiji’s macro text layout environment panel (Plugins -> New -> Macro):

// Comprehensive Cnidarian Planulae Larvae Morphometrics Pipeline
// Production Version 3.4 - Live Saving Log Display & Ingestion Invariant Resuming

macro "Compile Larvae Metrics Live View" {
    // 1. MASTER PROJECT DIRECTORY CONFIGURATION
    inputDir = getDirectory("Select Folder Containing Raw Larvae .oir Scans");
    if (lengthOf(inputDir) == 0) exit();
    
    outputDir = getDirectory("Select Local Target Folder for Result Assets");
    if (lengthOf(outputDir) == 0) exit();

    list = getFileList(inputDir);
    outputFile = outputDir + "Larvae_Morphometrics_Report.csv";
    
    // 2. CSV DATABASE STORAGE INITIALIZATION
    if (!File.exists(outputFile)) {
        File.saveString("File_Name,Ecto_L1,Ecto_L2,Ecto_L3,Ecto_L4,Ecto_L5,Ecto_Avg,Total_Area_um2,Gastro_Cavity_Area_um2\n", outputFile);
    }
    
    run("Clear Results");
    
    // Launch persistent Fiji terminal console to mirror updates live on display windows
    logWindowTitle = "[Live Larvae Morphometrics Log]";
    if (isOpen("Live Larvae Morphometrics Log")) {
        print(logWindowTitle, "\\Update:"); 
    } else {
        run("Text Window...", "name=" + logWindowTitle + " width=90 height=20 menu");
    }
    
    // Populate terminal log window instance directly with file history records from disk
    historicalText = File.openAsString(outputFile);
    print(logWindowTitle, historicalText);
    
    newProcessedCount = 0;
    alreadyCompletedCount = 0;
    
    // 3. CORE PROCESSING FOLDER LOOP INTERACTION
    for (i = 0; i < list.length; i++) {
        // Case-insensitive verification scanning across target extensions
        if (endsWith(list[i], ".oir") || endsWith(list[i], ".OIR")) {
            
            // INDEPENDENT BATCH RUN SEQUENCE AUTO-RESUME CHECK
            existingDatabaseText = File.openAsString(outputFile);
            if (indexOf(existingDatabaseText, list[i]) >= 0) {
                alreadyCompletedCount++;
                continue; // String entry verified! Bypass this sample file cleanly
            }
            
            newProcessedCount++; 
            
            // Open full-resolution object series using Bio-Formats core options
            run("Bio-Formats Importer", "open=[" + inputDir + list[i] + "] autoscale color_mode=Default view=Hyperstack stack_order=XYCZT");
            imgName = getTitle();
            
            run("ROI Manager...");
            roiManager("reset");
            
            // --- Part A: 5 Randomized Peripheral Ectoderm Stratum Thickness Lines ---
            ectLengths = newArray(5);
            sumEct = 0;
            setTool("line"); 
            for (j = 0; j < 5; j++) {
                lineNum = j + 1;
                waitForUser("Ectoderm Line " + lineNum + "/5", "Place a straight length ruler line across the outer ECTODERM layer.\nClick OK.");
                run("Measure");
                ectLengths[j] = getResult("Length", nResults - 1);
                sumEct += ectLengths[j];
                roiManager("Add");
                roiManager("Select", roiManager("count") - 1);
                roiManager("Rename", "Ecto_L" + lineNum);
            }
            avgEct = sumEct / 5;
            
            // --- Part B: Whole Cross-Sectional Tissue Envelope Perimeter Polygon Area ---
            setTool("polygon");
            waitForUser("Total Outer Boundary Contour", "Trace the outermost margin of the ciliated epithelial fringe perimeter.\nDouble-click your starting pixel node to close the loop, then click OK.");
            run("Measure");
            totalArea = getResult("Area", nResults - 1);
            roiManager("Add");
            roiManager("Select", roiManager("count") - 1);
            roiManager("Rename", "Total_Area");
            
            // UI SANITIZATION LAYERS: Clear frame states before inner loop entry
            roiManager("show none"); 
            run("Select None");
            setTool("polygon"); 
            
            // --- Part C: Internal Lumen Gastrovascular Cavity Polygon Area ---
            waitForUser("Gastrovascular Internal Cavity Outline", "Trace the inner open lumen border where the gastrodermal cells stop.\nDouble-click your starting pixel node to close the loop, then click OK.");
            run("Measure");
            cavityArea = getResult("Area", nResults - 1);
            roiManager("Add");
            roiManager("Select", roiManager("count") - 1);
            roiManager("Rename", "Cavity_Area");
            
            // Safely disengage target coordinate focus vectors
            run("Select None");
            roiManager("deselect");
            
            // --- Part D: Flatten Overlays & Burn Verified Peer-Review Image Audit ---
            roiManager("Show All with labels");
            run("Flatten");
            
            cleanBaseName = list[i];
            cleanBaseName = replace(cleanBaseName, ".oir", "");
            cleanBaseName = replace(cleanBaseName, ".OIR", "");
            saveAs("Jpeg", outputDir + cleanBaseName + "_Proof.jpg");
            close(); // Shuts down flat visual copy panel array frame securely
            
            // --- Part E: Live Streaming Row Appending & Live Log Echo Output ---
            rowString = list[i] + "," + ectLengths[0] + "," + ectLengths[1] + "," + ectLengths[2] + "," + ectLengths[3] + "," + ectLengths[4] + "," + d2s(avgEct,4) + "," + d2s(totalArea,2) + "," + d2s(cavityArea,2);
            File.append(rowString + "\n", outputFile); // Pushes database updates live to hard drive
            print(logWindowTitle, rowString);          // Updates terminal console pane panel layout
            
            run("Clear Results");
            close(); // Closes original raw data stack window frame to free workstation RAM
        }
    }
    
    roiManager("reset");
    showMessage("Ingestion Routine Complete!", "Batch run finalized successfully.\n\n• New profiles tracked: " + newProcessedCount + "\n• Continued elements skipped: " + alreadyCompletedCount);
}

4. Output Data Schema Matrix Architecture

The database schema compiled inside Larvae_Morphometrics_Report.csv follows a balanced transactional layout to ensure downstream compatibility with data-science toolkits (e.g., Python Pandas, R, or JMP):

Variable Suffix Header	Physical Measurement Units	Biological Matrix Definition
File_Name	String Text Field	Original identifier source microscope metadata file reference key.
Ecto_L1 to Ecto_L5	Micrometers ($\mu m$)	Raw single-width straight-line measurements across the ectoderm layer.
Ecto_Avg	Micrometers ($\mu m$)	Calculated true mathematical thickness mean per larval individual.
Total_Area_um2	Square Micrometers ($\mu m^2$)	Cross-sectional area of the entire external ciliated epidermal envelope plane.
Gastro_Cavity_Area_um2	Square Micrometers ($\mu m^2$)	Area of the internal open gastrovascular lumen cavity space boundaries.

4.1 Downstream Multivariate Analytical Equations

When importing this data matrix into statistical computing environments, you can automatically expand your phenotypic profiles by computing the following derived structural markers:

• True Collective Cellular Mass Area ($A_M$): Isorates absolute soft-tissue biomass, independent of structural distortions or water-inflation variations. \(A_M = \text{Total\_Area\_um2} - \text{Gastro\_Cavity\_Area\_um2}\)

• Proportional Ectodermal Investment Coefficient ($P_E$): Normalizes the raw peripheral wall thickness directly against overall body scale. \(P_E = \frac{\text{Ecto\_Avg} \times \sqrt{\text{Total\_Area\_um2}}}{\text{Total\_Area\_um2}}\)

• Lumen Volumetric Allocation Ratio ($R_S$): Tracks internal hydrostatic space allocation vs. somatic cellular building investment across phenotypes. \(R_S = \frac{\text{Gastro\_Cavity\_Area\_um2}}{\text{Total\_Area\_um2}}\)

---

5. Workstation Operational Instructions & Field Manual

To maximize digitization throughput, preserve experimental reproducibility, and prevent human click-fatigue across large multi-day imaging runs of the 28 planulae replicates, strictly follow these workspace mechanics:

• Global Environmental Scale Verification: Before executing a batch macro loop, open a single .oir file manually using the Bio-Formats plugin. Navigate to Analyze -> Set Scale.... Verify that the pixel-to-micron ratio has read cleanly from the microscope metadata, and click Global to ensure the coordinate matrix automatically locks onto every downstream slide asset change.
• Fiji Canvas Panning Shortcut: Tracing high-precision node loops with the polygon lasso tool requires deep magnification. To move across a magnified slice window smoothly without breaking your active selection path or dropping a rogue vector vertex, hold down the Spacebar. This temporarily toggles your mouse cursor into a grab-hand panning tool. Releasing the spacebar returns you instantly to your drawing tool.
• The Interface “Sticky” Selection Escape Trap: If Fiji’s interface becomes unstable or flags node vertices from a historic outer layer when you are trying to initiate an inner cavity trace loop, tap the Esc key on your keyboard. This serves as an immediate programmatic override, wiping active target hooks and freeing your cursor crosshairs to drop fresh node arrays cleanly.

• Cloud Ingestion Lock Defense: Background drive syncing triggers (e.g., Google Shared Drives) actively lock documents during network file validation, resulting in severe java.io.FileNotFoundException script crashes. Always declare your target Output Folder to a local hard drive sector (e.g., C:\Users\Name\Desktop\Scratch_Output). Once the macro pass finishes all 28 file entries and throws the final confirmation window, drag and drop the finished .csv and all matching _Proof.jpg image assets onto the shared cloud storage drive in a single, safe batch pass.