Abstract PO-021: TCGA differentially expressed genes between Caucasians and African Americans at variable analytic stringencies
Bibliographic record
Abstract
Abstract Introduction: In interrogating large dimensional datasets, many methods have evolved to correctly identify differentially expressed genes at acceptable false discovery rates. We examined the NCI TCGA head and neck gene set with open source programs (at varying stringencie)s between Caucasian and African-Americans (AA) specimens and catalogued differentially expressed genes. Procedures: cBioPortal and UALCAN were utilized to investigate the Tumor Cancer Genome Atlas (TCGA). We compared AA and Caucasian samples for this analysis. Tumor samples for both groups were controlled for tumor stage (T3 and T4), HPV status (-), and age (40-75). Three different methods were utilized to identify potential genes at acceptable high stringency (defined as probability <2 false positives out of 20,000 genes being tested). First, those genes differentially expressed in AA vs Cau populations, as well as, those significantly differentially expressed in lymph node-positive (N+) vs negative (N0) patients were investigated (p<0.01 in both groups). Secondly, Benjamin-Hochberg (BH) analysis analyzed differentially expressed genes between just AA vs Cau groups (q<0.05). Finally, genes between these two groups possessing a p-value <0.0001, but not included in the BH analysis, were included in a third analysis. Results: Benjamin-Hochberg analysis identified four differentially expressed genes (q<0.05). There were 15 additional genes in this groups with p-values <0.0001. Analysis of genes differentially expressed in both AA N+ vs N0, as well as, those differentially expressed in AA vs Cau (p<0.01) identified one additional gene (PPIL2). Expression of this gene was found to be increased in a step-wise fashion with increased tumor grade (p<0.05). Three of the four genes identified by BH analysis were differentially expressed in HNSCC tumors vs control (LIN52, CTNNA2, PRMT6). Increased expression of LIN52 and PRMT6 were grade dependent (p<0.05). In the third analysis (AA vs Cau differentially expressed genes with p<0.0001), seven of 15 genes were overexpressed in tumor vs normal tissue; five of these were associated with increased tumor grade (p<0.05). All genes associated with increased tumor grade were demonstrated increased expression in AA groups vs Cau. Conclusions: We conclude open source TCGA programs could be readily employed to identify differentially expressed genes between both groups at varying stringencies. The high-stringency Benjamin-Hochberg analytical tool identified 4 genes differentially expressed between AA vs Cau T3/T4 HPV- tumors. Genes differentially expressed in both AA N+ vs N0, as well as, AA vs Cau revealed one potential gene of interest. When applying a p-value of < 0.0001, 15 additional genes were identified. The low overall number of AA specimens in this dataset was limiting and future biobanks should favor enriched collection of all stages of head and neck cancer in under represented groups. Citation Format: Seth Buryska, Jacob Tupa, Frank Ondrey. TCGA differentially expressed genes between Caucasians and African Americans at variable analytic stringencies [abstract]. In: Proceedings of the AACR-AHNS Head and Neck Cancer Conference: Innovating through Basic, Clinical, and Translational Research; 2023 Jul 7-8; Montreal, QC, Canada. Philadelphia (PA): AACR; Clin Cancer Res 2023;29(18_Suppl):Abstract nr PO-021.
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 | 0.002 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.005 | 0.000 |
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".