PNNL-Comp-Mass-Spec / Kaiko_Publication

all the materials for the publication about the Kaiko

Kaiko_publication

All the data and materials for the publication about the Kaiko, which is a deep learning-based de novo peptide sequencing tool for natural bacterial isolates. We have deposited here both the data and the code for various figures in the manuscript.

• Figure 1 - Bacteria represented in training and testing data.

• Figure 2 - Accuracy of spectrum annotation.

• Figure 3 - Sequence prediction accuracy by peptide length.

• Figure 4 - Longest correct substring.

• Figure 5 - Taxonomic classification of natural isolates via proteomics.

• Supplemental Figure 1. Improving deep neural networks with more training data.

• Supplemental Figure 2 - Training and validation error.

• Supplemental Figure 3 - Distribution of peptide lengths for the training and testing dataset.

Figure 1

We generated a phylogenetic tree of the Kaiko benchmark datasets using graphlan, a python package. Please refer to this link for more information. You can see the annotation file and the tree input xml file in the phylo_vis folder. To tie the annotation file to the input tree,

$ python python graphlan_annotate.py --annot ../rplB.Tree/annotation_1_tree.txt ../rplB.Tree/rplB_0.xml

To generate the output image in a png format with setting up a resolution with --dpi and a size with --size (in inches),

$ python graphlan.py ../rplB.Tree/rplB_0.xml test.png --dpi 250 --size 7

Figure 2, 3, 4

Please refer to analysis/figures.ipynb.

Figure 5

To visualize the output results of taxonomy identification, we used the ETE Toolkit, which is another tool for analysis and visualization of any type of trees. Please refer to this link for more information. You can reproduce this from this notebook taxon_identification/fig5.ipynb.

Supplemental Figures

Please refer to analysis/suppl_figures.ipynb.

Contacts

Written by Joon-Yong Lee for the Department of Energy (PNNL, Richland, WA)
E-mail: [e-mail]
Website: https://panomics.pnl.gov/ or https://omics.pnl.gov

License

Licensed under the Apache License, Version 2.0; you may not use this file except in compliance with the License. You may obtain a copy of the License at https://opensource.org/licenses/Apache-2.0