Antônio Camargo
Antônio Camargo
`taxopy.TaxDb` should be able to download specific snapshots from the [`taxdump_archive`](https://ftp.ncbi.nlm.nih.gov/pub/taxonomy/taxdump_archive/).
Currently, users need to download one of the two model files (or train their own) before executing RNAsamba. Packaging pre-trained models in the library would spare users from downloading external...
Currently, RNAsamba only evaluates the transcript sequences that are given as input. But in some cases the user may not know the true orientation of the transcript, so a `--check_reverse_complement`...
As reported in previous issues (https://github.com/apcamargo/genomad/issues/78#issuecomment-1978551470 and https://github.com/apcamargo/genomad/issues/83), geNomad is currently incompatible with Keras 3. Adding support to Keras 3 would make the codebase more future-proof and potentially allow for...
Include functionality to export GFF and GBK files, as per [Issue #28](https://github.com/apcamargo/genomad/issues/28). This can be achieved via parsing the genes tabular output or by leveraging Pyrodigal's [`write_gff`](https://pyrodigal.readthedocs.io/en/stable/api/genes.html#pyrodigal.Genes.write_gff) and [`write_genbank`](https://pyrodigal.readthedocs.io/en/stable/api/genes.html#pyrodigal.Genes.write_genbank) methods....
Prediction intervals could be provided using conformal prediction methods. Adding prediction intervals to geNomad would allow users to better evaluate how confident they can be on specific classification instances.
When the output of `annotate` changes (due to a change in the sensitivity of the search, for instance), the `marker-classification` and `nn-classification` will skill skip some steps. In `marker-classification` this...
In response to [Issue #23](https://github.com/apcamargo/genomad/issues/23), geNomad's post-classification filters should be refined to take into account the circularity/completeness of input sequences. This could prevent the erroneous exclusion of legitimate plasmids due...
Gene strand can be very useful to detect prophages, but it is currently lacking from the `.gb` files. Because of that, there's no way to benchmark a tool that leverages...
Hey Rob, The *Paracoccus sanguinis* 5503 genome has a couple of duplicated protein IDs, which can cause troubles in downstream analysis. Here's the list: ``` WP_036707377.1 WP_036703025.1 WP_036705789.1 WP_036712647.1 WP_036705789.1...