Implementation of false positive DC correction and contaminated Poisson method
Bibliographic record
Abstract
The C++ code, spreadsheet, and example data contained in this archive accompany the manuscript: Shirley BC, Knoll JHM, Moquet J, Ainsbury E, Duy PN, Norton F, Wilkins RC, and Rogan PK. Estimating partial body ionizing radiation exposure by automated cytogenetic biodosimetry, submitted. ===== PartialBodyCalc.cpp (C++ source code) ===== PartialBodyCalc.cpp must first be compiled before execution. GSL and Boost libraries are utilized and must be installed before compilation. Usage: ./partialBodyCalculator CurvesFilePath TestCasesFilePath Execute software using provided example data: ./partialBodyCalculator ./Example_Curves ./Example_TestCases Short Description This software examines sample and curve data and computes estimated partial-body dose and estimated fraction of cells exposed. Curve File Content Example line: CNL_curve_automatedCurveFit_model208_Sigma14, 0.013006, 0.015516, 0.089047 Each line represents a curve and must consist of 4 comma-separated fields. Note: Fields 2, 3, and 4 can be found in ADCI by clicking on a curve within the curve list in the main window. Dose^2 is the quadratic coefficient, Dose^1 is the linear coefficient. Field 1: Curve Name (This name is also used in the TestCases file) Field 2: Quadratic Coefficient (Dose^2) Field 3: Linear Coefficient (Dose^1) Field 4: Intercept TestCases File Content Example line: CNL0and3x3 automatedCurveFit, 1679:298:16:6:1, 352, CNL_curve_automatedCurveFit_model208_Sigma14, 3.5 Each line represents a test case and must consist of 5 comma-separated fields. Field 1: TestCase name (Can be any string, only used in output to identify the TestCase) Field 2: Colon separated list of dicentric chromosome count bins. These values can be found in within the ADCI console after plotting the Poisson distribution of a sample. Do not include any additional spaces before or after each colon. Bins "1679:298:16:6:1" corresponds to 1679 images containing no dicentrics, 298 images containing 1 dicentric, 16 images containing 2 dicentrics, etc. Any unspecified bins are assumed to be 0. For example "1000:10" is equivalent to "1000:10:0:0:0:..." Field 3: True positive count (For TestCases which do not take into account a false positive adjustment, the true positive count should simply be the total dicentrics identified by ADCI. ie: "1679:298:16:6:1" in field 2 contains a total of 352 dicentrics. The total dicentric count can also be found in ADCI by clicking on a sample in the sample list within the main window, and locating the DC count for an appropriate sigma value in the console. If this TestCase is adjusted for FPs, enter the calculated true positive dicentric count here. Read more about FP adjustments below) Field 4: Curve name (This curve name must match a curve name specified in the curve file) Field 5: Dose in Gy at which 37% of irradiated cells survive. This depends on radiation type. Recommendation: 3.8 for X-Rays and 3.5 for gamma rays. False Positive Adjustment To create a TestCase which is adjusted for FPs, first you must calculate a true positive DC count for all calibration samples (see "CalibrationCurvesAndSamplesAdjustedForFalsePositives.xlsx" section below). After recalculating the DC frequency for each calibration sample using this true positive DC count, create a new (adjusted) calibration curve using the new frequencies. Note when using a curve adjusted for FPs to perform partial-body calculation, dicentric count bins specified in the TestCase do not change from the unadjusted TestCase, this program will remove the FP dicentrics before any comparison to the adjusted curve takes place. The only necessary changes to the TestCase are the true positive DC count and curve specified. ===== CalibrationCurvesAndSamplesAdjustedForFalsePositives.xlsx ===== The worksheet contains sheets labelled "Calibration", "WholeBody", and "Synthetic" for each of the four laboratories in the study. Note all formulae necessary to generate calibration DC frequencies adjusted for FP and the TP count of test samples are present in the appropriate cells. Of particular note are the "TP DC Count" and "Adjusted DC Frequency" columns in "Calibration" sheets and the "Bins" and "TP DC Count" columns in "WholeBody" and "Synthetic" sheets. Calculations assume a baseline DC frequency of 0.00078. Calibration curves adjusted for FP are built using values found in the "Adjusted DC Frequency" column. The "Bins" and "TP DC Count" fields are used the populate the TestCases file utilized by the PartialBodyCalc.cpp software.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.034 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.006 | 0.003 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.062 | 0.026 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".