M. R. Dwyer · Updated 31 October 2026
We connect accessible teaching material with the source records, measurements and comparisons that researchers can inspect. Species discovery begins with careful specimens, diagnostic characters and defensible comparisons. The collection distinguishes documented observations, exploratory calculations and proposed studies. The two proposed pilots describe future work and report no results.
Author-reviewed technical and educational reports, with source-linked data and figures. Independent computational checks passed. These reports retain their exploratory scope and do not establish clinical diagnostic validation or a new species determination.
Technical / exploratory reportSEPS:PRISM methods for specimen linked parasite morphology
We specify a reproducible route from specimen images to defensible morphological claims. SEPS:PRISM separates acquisition, boundary review, diagnostic measurement, candidate comparison and expert interpretation. We define the mathematical quantities that support conventional measurements, polygon descriptors, elliptic Fourier analysis and landmark shape analysis, then connect each quantity to calibration, tracing and sampling assumptions. The accompanying empirical analyses use recovered host counts, two photographed nematode profiles and published pentastome landmarks. They demonstrate specific computations rather than validated automated identification. We treat image readability, measurement accuracy and taxonomic adequacy as separate outcomes. This distinction allows useful observations to survive incomplete records while preventing a polished derivative from supplying evidence that the specimen never provided.
PDFDOCXHTML
SEPS FUZZARIA Evidence and Claim Limits Final · PDF
SEPS HookFuse Morphometric Methods Final · PDF
SEPS TEGUMENT Source and Boundary Review Final · PDF
Technical / exploratory reportSource linked morphometry of two nematode image profiles
We recover two visible rim profiles from one brightfield nematode photograph and compute measurements from saved source-coordinate polygons. The profiles have areas of 2856.2 and 3292.88 squared source pixels, maximum Feret diameters of 68.84 and 73.54 pixels, and circularities of 0.9631 and 0.9766. We then calculate elliptic Fourier coefficients directly from the actual traces and quantify reconstruction error over 512 perimeter positions. At eight harmonics, root mean squared reconstruction error reaches approximately 0.097 and 0.116 source pixels. These small errors describe the selected polygons rather than anatomical accuracy. We retain neutral region labels, explicit tracing provenance and missing physical calibration. The study supplies a reproducible geometric replacement for ambiguous legacy figures while reserving taxonomic and biological comparisons for independently reviewed, comparable specimens.
PDFDOCXHTML
Data and tables
Fourier_Coefficients.csv
Fourier_Reconstruction_Error.csv
evidence_record.json
exact_duplicate_groups.json
image_inventory.json
normalized_outline_coordinates.csv
radial_sensitivity.csv
registration_candidates.json
rim_metrics.csv
rim_vertices_source_coordinates.csv
SEPS_Nematode_Profile_Morphometry_Table_01.csv
Freshwater snail trematodes: linking form and identity
This proposed pilot would build a small reference collection of larval trematodes from freshwater snails using suitable preserved material. Microscopy and specimen records would document recognizable forms. Selected DNA markers would then test identification where preservation permits. The intended result is an illustrated reference that connects each identification to its specimen, images and supporting evidence.
Read pilot outline
Proposed pilot · No results reportedMealworm gregarines: shape, stage and identity
This proposed pilot would examine naturally occurring gregarines in commercially reared mealworms. It would ask how parasite shape and developmental stage relate to molecular identity. Photographs and measurements would be linked to selected DNA sequences where feasible. The intended outcome is a reference atlas that documents variation and supports careful identification of these overlooked invertebrate parasites.
Read pilot outline
Technical / exploratory reportMicroscopy evidence and inference in Cyclospora research
We examine how a microscopy reference can support recognition without implying viability, infectivity or validated automated diagnosis. We reconstruct the five-reader comparison reported by Ma and colleagues and pair it with authenticated CDC DPDx images. Across five readers, ultraviolet fluorescence agreement ranges from 74% to 94%, whereas modified acid-fast agreement ranges from 80% to 90% on the same 50-specimen set. Reader dependence, shared specimens and differences among biological endpoints prevent a single accuracy value from summarizing the problem. We specify source-linked image records and a prospective paired evaluation within SEPS:PRISM. The resulting gallery and reader comparison support education and study design; they do not establish clinical performance of SEPS.
PDFDOCXHTML
SEPS Cyclospora Comparative Microscopy Final · PDF
Data and tables
Published_Reader_All_Categories.csv
Published_Reader_Study.csv
SEPS PARAPLEX analysis of recorded reptile parasite assemblages
We apply the PARAPLEX ecological layer of SEPS:PRISM to 194 reptile observation rows in the recovered necropsy master workbook. The rows identify 81 hosts. Seventy-three hosts have positive parasite-group records and eight have explicit none entries. Forty-seven hosts have records in at least two broad groups. We retain unknown group cells and do not estimate population prevalence. The pentastome subtotal comprises 33 recorded organisms across 20 Basiliscus vittatus hosts, two across two Anolis equestris hosts and three in one Thamnophis sauritus. The first two components reconcile the 35-organism positive lizard summary at aggregate level. We preserve five repeated sample keys and separate 12 follow-up identification rows from new examinations. Five follow-up rows have candidates with agreeing host labels and counts; two have host-label conflicts, three have count mismatches and two lack a sample-key match. The resulting figures and tables support reproducible descriptions of the archive while identifying exactly which links and observation fields constrain ecological inference.
PDFDOCXHTML
SEPS PARAPLEX Recorded Assemblages Final · PDF
Data and tables
Analysis_Summary.json
Documented_Codetection_Counts.csv
Host_Summary.csv
Host_by_Recorded_Group.csv
Identification_Followup_Reconciliation.csv
Label_Crosswalk.csv
Recorded_Group_Summary.csv
Reptile_Observation_Ledger.csv
SEPS_PARAPLEX_Necropsy_Data.xlsx
Sample_Key_Collisions.csv
Source_Summary_Verification.json
Source_Rows_Verbatim.csv
Temporal_Record_Coverage.csv
Workbook_Inputs.json
SEPS_PARAPLEX_Necropsy_Assemblages_Table_01.csv
research figure; boundary review pendingN01 Source Linked Rim Traces
Source: 2023BaViNematode9(3).jpg. We show the full frame and enlarged crops with provisional polygon traces of two apparent bright rims. We preserve the source color and embedded microscope pointer. Cropping and display resizing constitute the only image operations. R1 and R2 do not imply separate specimens, homologous landmarks or a taxonomic determination. No physical scale is available.
PNGSVGPDFCaption
research figure; boundary review pendingN02 Rim Size and Shape
We recompute polygon area, Feret dimensions, circularity and Feret ratio from saved source-coordinate traces. Whiskers show a declared uniform radial displacement of ±2 source pixels; they are sensitivity ranges, not confidence intervals. R1 and R2 share one image and do not supply biological replication.
PNGSVGPDFCaption
research figure; boundary review pendingN03 Fourier Reconstruction and Error
We calculate elliptic Fourier coefficients from the piecewise-linear, constant-speed polygon parameterization and reconstruct 512 equally spaced perimeter positions. Root mean squared Euclidean error compares each reconstruction with the saved polygon. H = 32 retains more detail than H = 2 but does not create independent anatomical evidence. Source photograph pixels remain unchanged.
PNGSVGPDFCaption
research figure; boundary review pendingN04 Area Normalized Rim Overlay
We center each polygon at its area centroid, rotate it to the principal axis of 256 equally spaced boundary samples, and divide both coordinates by the square root of polygon area. The resulting profiles each have unit area. We do not treat their vertices as homologous landmarks or use this display as a taxonomic test.
PNGSVGPDFCaption
public literature figureC01 Published Reader Variation
We redraw counts from Ma et al. (2025), Table 1. The set contains 35 Cyclospora-positive, five Cystoisospora-positive and ten negative specimens. Repeated reader evaluations do not increase the number of independent specimens from 50 to 250. The right panel uses the source table denominator of 50 evaluations per reader; it is not a false-negative rate conditional on infection.
PNGSVGPDFCaption
reference gallery; retain CDC creditsC02 CDC Microscopy Gallery
CDC DPDx (2019), Cyclosporiasis gallery. We preserve source pixels and source arrows, arrange panels, and resize them for display. In panel F, the yellow arrow marks Cyclospora cayetanensis and the red arrow marks Cryptosporidium parvum. Panels show different preparations and specimens; they do not constitute repeated measurements of one oocyst. We infer no physical scale from display size or source magnification.
PNGSVGPDFCaption
public methods figureM02 Observation and Interpretation Ledgers
The observation ledger retains source-linked measurements. The interpretation ledger connects each conclusion to its evidence, alternatives and reviewer. Multiple analyses of one source image do not become independent corroboration. This design implements documentation rules rather than a numerical confidence score.
PNGSVGPDFCaption
public methods figureM03 Claim Specific Quality Gates
This proposed gate matrix illustrates task dependence and requires specialist tailoring. A taxonomic task that depends on an absolute size criterion also requires physical calibration. A qualitative description still requires source identity and reviewer agreement about what is visible. We do not interpret this matrix as an empirically validated FUZZARIA scoring rule.
PNGSVGPDFCaption
public methods figureM04 Biological Unit Partitioning
We define partition boundaries at the unit required by the intended generalization claim. Repeated views of one specimen never cross partitions. A source-disjoint evaluation further holds out acquisition sources, and grouped host or site tests address their respective dependencies. This figure describes evaluation design, not achieved accuracy.
PNGSVGPDFCaption
public methods figureM05 Separate Validation Endpoints
We evaluate detection, measurement, shape repeatability and taxonomic identification as separate endpoints. Human-reviewed references and appropriate units of independence govern each endpoint. A mask overlap statistic does not substitute for a diagnostic-character assessment.
PNGSVGPDFCaption
public attributed referenceS01 Raillietiella SEM Source Plate
Published Raillietiella orientalis image from Sapion-Miranda et al. (2025), Figure 3, recovered from the supplied ESA presentation. AH denotes anterior hook, PH posterior hook, and PA cephalic papilla. We preserve source labels and pixels and apply only proportional resizing and SEPS page layout. This is comparative literature material, not a SEPS specimen. License: CC BY 4.0.
PNGSVGPDFCaption
research measurement with anatomical review pendingS02 Projected Distance Sensitivity
SEPS calculation from retained coordinates on Shamsi et al. (2020), Figure 2 panel F. The selected endpoints span 192.0521 pixels. A 140-pixel bar labeled 100 µm gives 137.1801 µm. Independent endpoint displacement radii of 3 pixels and bar-end displacements of 1 pixel yield scenario bounds of 131.0226–143.5160 µm. These are geometric sensitivity bounds, not confidence limits or validated anatomical dimensions. We do not reproduce an altered source montage.
PNGSVGPDFCaption
public methods figureM06 Evidence Bundle
A claim remains inspectable when every derivative points to its source. Identity: Host, collection event, lot, specimen and voucher links. Acquisition: Original image, checksum, preparation and optical context. Measurement: Source coordinates, scale, units, algorithm and uncertainty. Interpretation: Diagnostic characters, comparators, alternatives and reviewer. These are proposed documentation requirements, not measured system performance.
PNGSVGPDFCaption
public methods figureM07 Archive Plate Standard
Proposed archive plate specification under current SEPS:PRISM. Whole specimen: Unannotated overview with specimen identifier and verified scale. Diagnostic views: Separate calibrated structures with orientation and preparation. Annotation layer: Labels outside anatomy; every region linked to source coordinates. Caption and files: Authorship, acquisition, transformations, source license and claim status. These are proposed documentation requirements, not measured system performance.
PNGSVGPDFCaption
public methods figureM08 Fourier Workflow
A numerical reconstruction cannot validate the anatomical choice of boundary. Reviewed polygon: Preserve source coordinates, closure, traversal and excluded artifacts. Arc length: Parameterize segments by length and integrate coefficients exactly. Reconstruction: Declare harmonic count and measure error against the chosen polygon. Comparison: Declare normalization; test orientation and repeat tracing before biology. These are proposed documentation requirements, not measured system performance.
PNGSVGPDFCaption
public methods figureM09 Claim Caps
FUZZARIA records why a task can or cannot support a particular conclusion. Visible appearance: Describe a documented feature without assigning unsupported identity. Measured character: Require relevant boundary, scale, units and uncertainty. Taxonomic comparison: Use diagnostic anatomy and authenticated comparative material. Final determination: Retain specialist decision, contradictions and unresolved alternatives. These are proposed documentation requirements, not measured system performance.
PNGSVGPDFCaption
Technical methodologyM01 SEPS PRISM Architecture
SEPS:PRISM retains FieldView or direct import, TEGUMENT, HookFuse, task-specific candidate comparison, FUZZARIA and expert review. Evidence export preserves observations and interpretations separately. PARAPLEX adds downstream ecological analysis without replacing those components. This collection executes source ingestion, deterministic morphometry and descriptive ecological summaries; the diagram does not claim that every recognition or ecological model is implemented or validated. Each output retains its unit of observation, missing evidence and reporting scope.
PNGSVGPDFCaption
Technical methodologyC03 Reader Error Categories
We transcribe every reported outcome category in Ma et al. (2025), Table 1. Nine reader-method category totals equal 50. Reader 3 MAF sums to 49; the black segment displays the arithmetic residual to 50 and does not invent a clinical outcome. Discrepant assignments and additional-coccidian reports explain the other gaps left by agreement, false-negative and false-positive counts alone. Repeated evaluations share specimen identities. UV means ultraviolet wet-mount microscopy and MAF means modified acid-fast staining. The source percentages use all 50 specimens, not class-conditional denominators.
PNGSVGPDFCaption
Source-linked descriptive analysisP01 Host by Recorded Parasite Group
The matrix preserves three documentary states for 81 source-local reptile host identifiers. Blue records a positive source label. Orange records a host row that explicitly says none, without certifying group-specific examination completeness. Pale cells contain no record for that group and remain unknown in ecological inference. Groups aggregate source terminology and do not represent species. We apply no image post-effects.
PNGSVGPDFCaption
Source-linked descriptive analysisP02 Recorded Group Counts
Left counts distinct source-local hosts with at least one record in each broad group. Right sums explicitly numeric or unambiguously additive counts across source rows. The sums describe the ledger, not independently verified numbers of unique parasites. Approximate counts, an open quantity and a missing cyst count remain separate. Reused vial identifiers and overlapping recovery records may affect individual totals; the supplied collision table identifies those cases.
PNGSVGPDFCaption
Source-linked descriptive analysisP03 Documented Codetection Counts
Each off-diagonal cell counts host identifiers with positive source records in both groups; the diagonal counts hosts with that group. We compute C = B transpose B for a documentary positive-record indicator B. Unrecorded taxa do not supply biological negatives. We therefore estimate neither odds ratios nor transmission edges from this matrix. Repeated records within a host contribute only once to group presence.
PNGSVGPDFCaption
Source-linked descriptive analysisP04 Temporal Record Coverage
Year-specific counts derive from the date field in the main reptile section. The workbook contains no main-series records for 2020. NCX-H025 has six different date values within 2021, so we retain that conflict and use only its consistent calendar year here. The 2024 identification follow-up uses a different date meaning and does not enter this plot. No equal sampling effort, seasonality or population trend is assumed.
PNGSVGPDFCaption
Source-linked descriptive analysisP05 Identification Link Reconciliation
We normalize MBH and MBHarvey prefixes only to propose sample-key matches. We retain raw identifiers. A consistent candidate agrees on the normalized host spelling and recorded count; it does not establish a confirmed specimen link. Host-label conflicts, count mismatches and unmatched BaVi20 references remain unresolved. We do not transfer the reported M. ingens identification into the primary ledger or count the follow-up material again.
PNGSVGPDFCaption
