Homologous Gene Finding with a Hidden Markov Model
Bibliographic record
Abstract
The homology search problem and the gene finding problem are two fundamental problems in bioinformatics. The homology search problem is to find the homologous regions of two biological sequences; the gene finding problem is to find all the genes in both strands of a genomic sequence. Recently, gene finding research has demonstrated that homology search results can be used to improve the accuracy of gene finding. By combining the two problems, we define a new problem called the homologous gene finding problem. The homologous gene finding problem is to find homologous genes of a query gene in a target genomic sequence. \n \nConsequently, we present a new homologous gene finding algorithm in this thesis. We borrow the idea of gene mapping and alignment algorithms, and apply existing seed-based homology search algorithms and hidden Markov model-based (HMM-based) gene finding algorithms to solve the homologous gene finding problem. After we find high-scoring segment pairs (HSPs) between the query gene and the target genomic sequence, we locate target regions that we believe contain a gene homologous to the query gene. Then, we extend existing HMM-based gene finding algorithms to find homologous gene candidates. To improve the accuracy of homologous gene finding, we train a HMM to be biased toward the query gene. We also introduce a new coding sequence (CDS) length penalty as a measure of how the CDS lengths of the query gene and its homologous gene vary to further improve the accuracy. We use the new CDS length penalty together with our enhanced Viterbi algorithm and our flexible finish condition to improve the speed of homologous gene fining without harming the accuracy. Finally, we use protein alignment to pick and rank the best homologous gene candidates. \n \nIn this thesis, we also describe several experiments to evaluate and support our homologous gene finding algorithm.
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 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.003 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 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".