Seqera Platform Feature Requests

Platform currently supports parameters that are simple types (strings, numbers, etc), but Nextflow supports more complex parameters via the params-file argument. These can be arbitrarily nested maps and lists. Currently workflows that expect this type of parameter will not be able to take those parameters in Platform due to the restrictions on parameter type. It would be helpful if users could specify these parameters in Platform without removing the schema.json .

Can we add custom icon support to s3 compatible buckets in the Seqera UI? Essentially let the user upload a custom icon when adding credentials for the s3 compatible storage so that icon gets assigned to all buckets for that server base url.

To quickly find relevant dataset entries, it would be helpful to dynamically reduce the list-view to entries that match the filter key:value (for example, the username). So if I type username:ev that should dynamically filter the list-view to only show entries that were createdBy any Platform username that matches ev* (so, evan , evanfloden , eve , evangeline ).

The new datasets view now offers improved options for searching by specific elements. It would be helpful to extend the search to look through keywords in the description field and the samplesheet metadata itself. Often the description contains more detailed information and relevant keywords (like crisprseq ) that is not reflected in the title and appending this to the labels requires extra effort. Similarly, it can be interesting to find a specific sample that was processes in multiple pipelines across multiple individual Datasets, e.g. "search for SRR1234567890 and give me all datasets that contain this entry".

When launching a pipeline in Seqera Platform, the data selection for Data Explorer should filter for buckets with write permissions, especially when browsing for workdir or outdir .

Many users develop their Python or R notebooks, or Nextflow pipelines in VSCode, on remote EC2 instances. Once a Studio session is in a running state, users should be able to connect to the remote Studio session via a SSH tunnel to the EC2 instance with their existing VSCode/Vim/Bash environment(s). There is a VSCode plugin available that can be reviewed for inspiration.

Add support for connecting to Databricks sources and displaying data directly in Data Explorer.

When selecting a CE for during Studio creation, for example, I want to avoid selecting a Batch CE. Unfortunately that's only easy to do if the person who made the CE was good with their naming strategy. Could we have a batch vs cloud icon of some kind in CE drop-downs to facilitate this?

When launching larger pipelines with many parameters, it is possible to hit the limit imposed by Platform, giving the error: ParamsText field too big - Max allowed size 20480 characters When this limit is reached, it can become difficult to launch the pipeline. Is it possible to increase or remove this limit?

Currently, Data Explorer does not support previewing BAM or BED files. An in-platform, interactive experience can be provided by embedding the IGV Web App ( igv.js ) in Data Explorer. This allows in-browser previewing many common bioinformatic file types .