ProLink is a python script that allows to execute multiple proteomic analysis tools automatically.
This software is intended to be executed in Google Colab. To run it, open this notebook and follow these steps:
Step 1: Run the first cell in order to install the dependencies required. It is only needed to do once everytime the execution environment is initialized.
Step 2: Introduce the desired parameters in the form of the second cell and execute it.
Step 3: Run the "Execute the script" cell. This may take a while to be completed. Check the box extra_verbose for an enriched output beyond the standard.
Step 4: Download the results as a zip file by running the last cell.
Parameters
| Argument name | Description |
|---|---|
| query_proteins | Protein sequence code to query. Will be searched on NCBI. |
| hitlist_size | Number of found sequences to obtain after BLAST. |
| blast_database | Database used in BLAST. |
| pro_blast_ | Boolean to select Pro BLAST or regular BLAST. |
| length_restrict | Boolean to resctrict the length of the found sequences with respect the query. |
| length_margin | Number to multiply the query length and restrict the min and max length of the found sequences. |
| cluster_seqs | Boolean to select if clustering the sequences. |
| min_seq_id | Initial minimum sequence identity treshold to cluster together (0-1). |
| pro_clustering_ | Boolean to select Pro Clustering or regular clustering. |
| min_seq_id_step | Step to increase or decrease the minimum sequence identity threshold while Pro Clustering. |
| min_number_of_clusters_to_cluster_again | Minimum number of clusters allowed when using Pro Clustering. |
| max_number_of_clusters_to_cluster_again | Maximum number of clusters allowed when using Pro Clustering. |
| check_pfam_domains | Boolean to select if checking the Pfam domains of the sequences. |
| align_seqs | Boolean to select if aligning the sequences. |
| trim | Boolean to select if trimming the alignment retaining phylogenetically-informative sites. |
| generate_logo | Boolean to select if generating a sequence logo. |
| generate_tree | Boolean to select if generating a phylogenetic tree. |
| tree_type | Type of phylogenetic tree. Either "NJ" (Neighbor Joining) or "ML" (Maximum Likehood). |
Advanced parameters (in ProLink/parameters.yaml)
| Argument name | Description | Default value |
|---|---|---|
| max_low_identity_seqs | Maximum number of low identity seqs to find when using "Pro BLAST". | 1 |
| min_low_identity_seqs | Minimum number of low identity seqs to find when using "Pro BLAST". | 1 |
| expected_min_identity | Maximum identity percentage to consider a sequence a low identity seq when using "Pro BLAST". | 0.25 |
| additional_hits | Number of additional sequences to find when using "Pro BLAST". | 2000 |
| weblogo_format | Output format when using generate_logo. | 'png' |
| bootstrap_replications | Number of bootstrap replications when generating the tree. Needs to be 100, 250, 500, 1000 or 2000. | 100 |
| output_dir | Name of the outputs directory. | '' |
This software can also be installed locally in a Linux machine. To do so, installing it with the conda package manager is advised.
Use the file prolink_env.yaml to create an environment with almost all the required dependencies. Activate it afterwards. Aditionally, MEGA must be installed manually if you expect to generate trees.
conda env create -f https://raw.githubusercontent.com/unizar-flav/ProLink/main/prolink_env.yaml
conda activate prolinkTo run it locally, use the following command. For additional help on the usage: prolink --help
prolink [-f .yml] [--opt opt=val [opt=val ...]] [--verbose] QUERY_CODEPro BLAST allows to search for homologous sequences via the BLAST tool, but making sure that low identity seqs are also represented in the output.
Firstly, a regular BLAST is launched and hitlist_size seqs are obtained. If the number of low identity seqs in the found seqs are below the min_low_identity_seqs parameter, a new regular BLAST will be launched but with hitlist_size += additional_hits. It will stop when the number of low identity seqs reach max_low_identity_seqs.
Pro Clustering allows to obtain a number of clusters in a determined range, unlike regular clusterings that uses MMseqs2 to get an undetermined number of clusters based on a similarity value.
Firstly, a regular clustering with the determined minimum sequence identity is executed. If the number of clusters is avobe the max_number_of_clusters_to_cluster_again value, the sequences will be clustered again but with min_seq_id += min_seq_id_step, in order to obtain an inferior number of clusters. On the contrary, if the number of clusters is below the min_number_of_clusters_to_cluster_again, the min_seq_id requested will be decreased.
Chiu, J.K.H., Ong, R.TH. (2022). Clustering biological sequences with dynamic sequence similarity threshold. BMC Bioinformatics 23, 108.
Ondov, B. D., et al. (2016). Mash: fast genome and metagenome distance estimation using MinHash. Genome Biology, 17(1), 132.
Steinegger, M., Söding., J. (2017). MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nature Biotechnology, 35, 1026.
Edgar, R. C. (2004). MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research, 32(5), 1792-1797.
Crooks, G. E., et al. (2004). WebLogo: A sequence logo generator, Genome Research, 14:1188-1190.
Tamura, K., et al. (2021) MEGA11: Molecular Evolutionary Genetics Analysis version 11. Molecular Biology and Evolution 38:3022-3027.