Loading BioAtlas...
Loading BioAtlas...
将生物大模型、对接计算引擎与组学管线并排横向对比。按获取方式、硬件开销及细分领域灵活筛选，在一分钟内得出客观可辩护的技术选型方案。
| 技术参数指标 | ＋ 添加工具 | |||
|---|---|---|---|---|
| Category | Scientific ML Libraries & Supporting Tools | Scientific ML Libraries & Supporting Tools | Scientific ML Libraries & Supporting Tools | — |
| Subcategory | Scientific ML | Conventional or Statistical Baselines | Conventional or Statistical Baselines | — |
| Company | DeepGraphLearning / TorchDrug contributors | Research / method authors | Research / method authors | — |
| Access mode | Public code / project terms | Confirm distribution and exact release | Confirm distribution and exact release | — |
| Licence conditions | Review the linked release, model weights, datasets and service terms separately | NOT_INDEPENDENTLY_AUDITED | NOT_INDEPENDENTLY_AUDITED | — |
| Compute requirements | Consult linked installation guidance; workload dependent | Version and data dependent | Version and data dependent | — |
| Biological application | 未明确记录 | 未明确记录 | 未明确记录 | — |
| Research workflow | 未明确记录 | 未明确记录 | 未明确记录 | — |
| Evidence level | Project or provider documentation; no local performance benchmark | PEER_REVIEWED_METHOD_OR_CONFERENCE | PEER_REVIEWED_METHOD_OR_CONFERENCE | — |
| Commercial eligibility | Not independently confirmed | NOT_INDEPENDENTLY_AUDITED | NOT_INDEPENDENTLY_AUDITED | — |
| Price | See project terms; compute costs may apply | Verify access and licence | Verify access and licence | — |
| Availability ⓘ | See project access and maintenance status | Method paper or official documentation | Method paper or official documentation | — |
| Features | Machine-learning toolbox for molecular graphs and drug research, Input: Molecular or protein datasets and task definitions, Output: Trained models, representations and predictions | HMMER 3 scientific method, Query sequences or profile HMM and sequence database, Profile HMM hits and statistical homology significance | MMseqs2 scientific method, Amino acid queries and reference sequence database, Sequence matches, clusters and alignment results | — |
| Entry type | Research software / model | Classical/statistical scientific method | Classical/statistical scientific method | — |
| AI role | Scientific machine learning | NON_AI_CLASSICAL | NON_AI_CLASSICAL | — |
| Input data | Molecular or protein datasets and task definitions | Query sequences or profile HMM and sequence database | Amino acid queries and reference sequence database | — |
| Output data | Trained models, representations and predictions | Profile HMM hits and statistical homology significance | Sequence matches, clusters and alignment results | — |
| Validation | Evaluate on representative data; no clinical suitability established | Published method scope; no independent BioAtlas execution or benchmark | Published method scope; no independent BioAtlas execution or benchmark | — |
| Type | 未明确记录 | 未明确记录 | 未明确记录 | — |
| Intended use | Biological and pharmaceutical research | 未明确记录 | 未明确记录 | — |
| Compatibility | 未明确记录 | 未明确记录 | 未明确记录 | — |
| Limitations | Framework compatibility and dataset quality require review; a library does not provide validated drug candidates. | 未明确记录 | 未明确记录 | — |