Functional assessments of <i>PTEN</i> variants using machine-assisted phenotype scoring
Bibliographic record
Abstract
ABSTRACT Genetic testing is widely used in evaluating a patient’s predisposition for developing a malignancy. In the case of cancer, when a functionally impactful inherited mutation (i.e. genetic variant) is identified in a disease-relevant gene, the patient is at elevated risk of developing a lesion in their lifetime. Unfortunately, as the rate and coverage of genetic testing has accelerated, our ability to make informed assessments regarding the functional status of the variants has lagged. Currently, there are two main strategies for assessing variant functions: in silico predictions or in vitro testing. The first approach is to build generalist computational prediction software using theoretical parameters such as amino acid conservation as feature inputs. These types of software can classify any variant of any gene. Although versatile, their non-specific design and theoretical assumptions result in different models frequently producing conflicting classifications. The second approach is to develop gene-specific assays. Although each assay is tailored to the physiological function of the gene, this approach requires significant investments. Therefore, there is an urgent need for more practical, streamlined and cost-effective methods. To directly address these issues, we designed a new approach of using alterations in protein subcellular localization as a key indicator of loss of function. Thus, new variants can be rapidly functionalized by using high-content microscopy. To facilitate the analysis of large amounts of image data, we developed a new software, named MAPS for machine-assisted phenotype scoring, that utilizes deep learning (DL) techniques to extract and classify cell-level phenotypes. This new Python-based toolkit helps users leverage commercial cloud-based DL services that are easy to train and deploy to fit varying experimental conditions. Model training is entirely code-free and can be done with limited number of images. Users simply input the trained endpoints into MAPS to accomplish cell detection, phenotype discovery and phenotype classification. Thus, MAPS allows cell biologists to easily apply DL to accelerate their image analysis workflow.
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) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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".