Paper · S7614
Deep Profiling of Mouse Splenic Architecture with CODEX Multiplexed Imaging
CODEX imaging maps cells and neighborhoods in nine mouse spleens. Most changing contacts track cell abundance rather than stronger inferred interactions. Neighborhood associations leave causal direction open. Image classification used test-set patches for tuning, and reported precision is not clinical accuracy. Financial interests and processing limits are retained.
Read the paper at ebi.ac.ukWhat we checked
Read101,413 XML characters in13 complete returns,20 references,158 physical table cells including15 blank separators, and15 meaningful JPEGs. Read the complete one-page S4 PDF in text and pixels:29 marker rows and58 numeric values. Nine female mice supplied three control and six disease-model spleens. Nolan’s roles include conceptualization, editing, supervision, administration and funding. Native August9 print date differs from body Published August2. Root also read1,085 native cells, including four blanks, in six S1/S2 workbook sheets through eight complete returns. Printed spleen counts total707,466, matching the post-filter main count. Native panel denominators and phenotype labels remain distinct.
Complete supplied main, figures, S1/S2 cells and S4 PDF
How we check sources
Not checked yet
- The interaction workbook, two videos, original main PDF, raw data, code, original workbook layouts and every fine figure glyph remain unreviewed. Interaction cells and formulas were inventoried mechanically with no reading or recalculation credit. Caption labels, cross-references, thresholds and date correspondence retain stated discrepancies.
- Cells, fields, patches and graph edges are nested within mice. No cohort randomization or blinding was used. Test-set images contributed to tuning;90% patch precision does not establish human diagnostic accuracy, causal mechanisms or UAP findings.
- Akoya ownership, consulting, advisory and employment disclosures and a Stanford patent are reported historical interests. Defense and Northrop funding do not establish a secret programme or government material access. Repeating22 markers three times does not demonstrate66 distinct antigens.
- S1/S2 clarify antibody versus channel counts and repeated marker panels. A cell-count phenotype label differs in spelling and sign from its annotation label; neither is silently corrected. Summing printed counts is not reproducing segmentation or adding independent subjects.
Cite this record
Yury Goltsev, Nikolay Samusik, Julia Kennedy-Darling, Salil Bhate, Matthew Hale, Gustavo Vazquez, Sarah Black, Garry P. Nolan. Deep Profiling of Mouse Splenic Architecture with CODEX Multiplexed Imaging. August 9, 2018. BioParity S7614. https://bioparity.io/sources/S7614Record data (JSON)