Add support for NetMHCpan-4.2#22
Open
yannic-chen wants to merge 6 commits into
Open
Conversation
NetMHCpan-4.2 shares the same command-line interface and tabular output format as 4.1, so the job construction and output parser work unchanged. This wires 4.2 through the config/auto-detection paths while keeping 4.1 as a supported fallback: - parameters.py: the "auto" NetMHCpan path now resolves to the bundled netMHCpan4.2 wrapper. - gui.py initialize(): accepts a netMHCpan-4.2 (preferred) or netMHCpan-4.1 (legacy) tools directory and downloads the matching data files. - gui.py download_data_file(): new netMHCpan4.2 branch pointing at the DTU NetMHCpan-4.2 data archive. - tool_scripts/netMHCpan4.2: bash wrapper mirroring the 4.1 one with NMHOME set to netMHCpan-4.2. - Updated GUI citation, docs and CHANGELOG; bumped version to 0.7.12. Co-Authored-By: Claude Opus 4.8 <[email protected]> Claude-Session: https://claude.ai/code/session_0165QaBkysJcvX2ZizxWyvop
Make a fresh install work on current Python and pip: - plotly_venn.py: import Iterable from collections.abc; the collections.Iterable alias was removed in Python 3.10. - requirements.txt / setup.py: add version ceilings matching the last releases MVP works with -- pandas<2.0 (DataFrame.append removed in pandas 2.0), numpy<2.0, dash<3.0 with dash-bootstrap-components 1.x (Dash 3.0 removed the standalone dash_*_components packages), and kaleido==0.2.1 / plotly<6. Also add seaborn and mhcnames to setup.py. Verified the CLI and GUI import chains build cleanly on Python 3.10 with this set (pandas 1.5.3, numpy 1.26.4, dash 2.18.2, dbc 1.7.1). Co-Authored-By: Claude Opus 4.8 <[email protected]> Claude-Session: https://claude.ai/code/session_0165QaBkysJcvX2ZizxWyvop
Replace the pandas DataFrame.append() call in cl_tools.make_binding_predictions with pd.concat() (append was removed in pandas 2.0). Empty frames are filtered out of the concat so the pandas>=2.1 empty/all-NA FutureWarning does not fire. This removes the only reason pandas/numpy were version-capped, so those pins are dropped. dash<3.0 (+ dash-bootstrap-components 1.x) is still required because the GUI imports the standalone dash_*_components packages that Dash 3.0 removed; that constraint is independent of the Python version. Verified on Python 3.12 with pandas 3.0.3 / numpy 2.5.1 / dash 2.18.2: full CLI and GUI import chain, the concat path (no FutureWarning), and the NetMHCpan output parser all work. Co-Authored-By: Claude Opus 4.8 <[email protected]> Claude-Session: https://claude.ai/code/session_0165QaBkysJcvX2ZizxWyvop
NetMHCIIpan changed its output column layout across 4.0->4.3, and MVP parsed results by fixed column position, so newer versions would be misparsed. - netmhcpan_helper: parse NetMHCpan/NetMHCIIpan output by column *header name* (Score_EL, %Rank_EL, Score_BA, Affinity(nM)/Aff(nM), %Rank_BA) instead of a fixed index. Positions are read from the results header each block, so added or reordered columns no longer break parsing. Falls back to the previous fixed indices if no header is found, so existing behavior is preserved. - gui.initialize(): accept netMHCIIpan-4.0 through 4.3 (newest first), mirroring the NetMHCpan handling. - gui.download_data_file(): derive the DTU service name and destination folder from the version, so any netMHCpan/netMHCIIpan version resolves automatically. - tool_scripts/netMHCIIpan: auto-detect the newest installed netMHCIIpan-4.x and dispatch to the right entry point (4.0 Perl frontend, or 4.1+ launcher/binary). - Docs/CHANGELOG updated. Verified on Python 3.12: parser maps columns correctly for a class I 4.2 header, a class II 4.0 header, a class II header with shifted columns (simulating 4.3), and the no-header fallback; full CLI and GUI imports still succeed. Co-Authored-By: Claude Opus 4.8 <[email protected]> Claude-Session: https://claude.ai/code/session_0165QaBkysJcvX2ZizxWyvop
- docs/INSTALL.md: step-by-step guide for installing from source and running MhcVizPipe interactively or as a server -- Python environment, DTU tool setup (NetMHCpan-4.2, NetMHCIIpan 4.0-4.3, GibbsCluster), configuration, a CLI smoke test, running the GUI server behind an SSH tunnel / systemd, and troubleshooting the errors commonly hit during setup. - README: link to the new guide from the Installation section. - test_data/class_I readme: correct the mouse alleles from H2-Kb/H2-Db to the recognized H-2-Kb/H-2-Db. - CHANGELOG: note the guide and the allele-name correction. Co-Authored-By: Claude Opus 4.8 <[email protected]> Claude-Session: https://claude.ai/code/session_0165QaBkysJcvX2ZizxWyvop
SVG app icon: a sequence-logo motif (Y and C as the tall anchor residues) with an 'MVP' wordmark and a '0.7.12' version badge to visually distinguish this build. Co-Authored-By: Claude Opus 4.8 <[email protected]> Claude-Session: https://claude.ai/code/session_0165QaBkysJcvX2ZizxWyvop
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
NetMHCpan-4.2 shares the same command-line interface and tabular output format as 4.1, so the job construction and output parser work unchanged. This wires 4.2 through the config/auto-detection paths while keeping 4.1 as a supported fallback:
Claude-Session: https://claude.ai/code/session_0165QaBkysJcvX2ZizxWyvop