Table 1.
Computational approaches to identify and quantify mRNA terminal ends in LRS data
| Approach | Known annotations | De novo identification | Terminal end selection | Orthogonal features | Full-length status | Reference |
|---|---|---|---|---|---|---|
| TAPIS | Optional | Clustering | Longest isoform with shared splice sites | Terminal end A's, terminal end adapters | No | (Abdel-Ghany et al. 2016) |
| StringTie2 | Optional | Graph-based | Unknown | No | No | (Kovaka et al. 2019) |
| FLAIR | Yes | Clustering | Read density | Terminal end adapters | Read level | (Tang et al. 2020) |
| TALON | Yes | Clustering | Clustering | Terminal end A's | Transcript level, FSM/ISM/NIC categories | (Wyman et al. 2020) |
| FLAMES | Yes | Clustering | Annotations | SRS RNA-seq | No | (Tian et al. 2021) |
| Bambu | Yes | Probabilistic ML model | Categorization and modeling | Terminal end A's | Read level and transcript level | (Chen et al. 2023) |
| ESPRESSO | Yes | Clustering | Splice junctions of terminal exons | No | Read level and transcript level with FSM/ISM/NIC categories | (Gao et al. 2023) |
| IsoQuant | Optional | Graph-based | Clustering | Terminal end A's | No | (Prjibelski et al. 2023) |
| IsoTools | Yes | Graph-based | Peak calling, read density | No | No | (Lienhard et al. 2023) |
| SQANTI3 | Recommended | Probabilistic ML model | Categorization and modeling | Terminal end A's, terminal end adapters, SRS RNA-seq, complementary SRS data sets | Transcript level, FSM/ISM/NIC categories | (Pardo-Palacios et al. 2024a) |
