A Hierarchical HMM Implementation for Vertebrate Gene Splice Site Prediction
Bibliographic record
Abstract
With the current volume of genomic information available, and the rate at which new data is being accumulated, there is a tremendous need for tools to effectively manage the information. It is here that bioinformatics attempts to begin solving problems. \nA common scenario occurs when a newly sequenced piece of genomic data is produced. There are several questions that are often asked about this piece of data. Is it similar to an existing piece of data? What are the coding regions of the sequence? The first question can be answered using a tool called BLAST (Basic Local Alignment Search Tool) which is capable of finding homologies to existing sequences. It is the latter question that we attempt to solve in this endeavor. \nIn a sequence of genomic data, a gene occurs as a band of alternating introns (noncoding) and exons (coding). At first the entire sequence is transcribed, but the non-coding regions are subsequently spliced out. The start of an intron is marked with a sequence of nucleotides called a donor site, and the end is marked with an acceptor site. The donor and acceptor sites are used mechanistically in the removal of the intron. What is left after the splicing is the raw coding sequences that will be used to build up the proteins. \nBy first identifying the donor and acceptor regions of a sequence, it is then known which regions will be spliced out of the transcribed sequence. We attempt to solve the problem by creating a tool that will predict the locations of these donors, acceptors and subsequent exon regions in a raw genomic sequence. \nDue to the transcriptional machinery, the donor and acceptor sites in a genomic sequence has stochastic signals or patterns which can be utilized for recognition. Similarly, exons or gene encoding areas also exhibit faint patterns which can be exploited for recognition.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.002 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.059 | 0.037 |
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".