Unsupervised analysis of multi-experiment transcriptomic patterns with SegRNA identifies unannotated transcripts
Bibliographic record
Abstract
Abstract Background Exploratory analysis of complex transcriptomic data presents multiple challenges. Many methods often rely on preexisting gene annotations, impeding identification and characterization of new transcripts. Even for a single cell type, comprehending the diversity of RNA species transcribed at each genomic region requires combining multiple datasets, each enriched for specific types of RNA. Currently, examining combinatorial patterns in these data requires time-consuming visual inspection using a genome browser. Method We developed a new segmentation and genome annotation (SAGA) method, SegRNA, that integrates data from multiple transcriptome profiling assays. SegRNA identifies recurring combinations of signals across multiple datasets measuring the abundance of transcribed RNAs. Using complementary techniques, SegRNA builds on the Segway SAGA framework by learning parameters from both the forward and reverse DNA strands. SegRNA’s unsupervised approach allows exploring patterns in these data without relying on pre-existing transcript models. Results We used SegRNA to generate the first unsupervised transcriptome annotation of the K562 chronic myeloid leukemia cell line, integrating multiple types of RNA data. Combining RNA-seq, CAGE, and PRO-seq experiments together captured a diverse population of RNAs throughout the genome. As expected, SegRNA annotated patterns associated with gene components such as promoters, exons, and introns. Additionally, we identified a pattern enriched for novel small RNAs transcribed within intergenic, intronic, and exonic regions. We applied SegRNA to FANTOM6 CAGE data characterizing 285 lncRNA knockdowns. Overall, SegRNA efficiently summarizes diverse multi-experiment data.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".