UFCG Tutorial

Follow this step-by-step guide with a small sample dataset to learn how to use UFCG pipeline.

Sample dataset

Sample species

Six representative fungal species and their whole genome assemblies are prepared as a sample dataset.

GCA_000091045.1

GCA_000146045.2

GCA_000149205.2

GCA_000182565.2

GCA_000271585.1

GCA_000697725.1

Sample files

After running following command, you will find all the required files from sample/ directory.

$ ufcg download -t sample

The directory contains following components.

• seq/ : Directory with the whole genome assemblies of the species above
• out/ : Directory containing previously extracted core gene profiles
• tree/ : Directory to store result phylogenetic trees
• meta_full.tsv : Metadata file of the species with full information
• meta_simple.tsv : Metadata file of the species with minimum information

Profile core gene UFCG profile

By following this guide, you will

• extract core genes from the genomes in directory /sample/seq/,
• and save their profile on /sample/out/ directory.

Run profile module

Launch profile module in interactive mode by entering following command on your terminal.

$ ufcg profile -u

The program will tell you to enter the input file or directory. We want to use the entire directory as an input, therefore you should enter the path where the genomes are lying.

UFCG |: Enter the file or directory ... (--input) : sample/seq/

Enter the output directory in the same way.

UFCG |: Enter the directory ... (--output) : sample/out/

We want to extract protein markers from the sequences. Type 'PRO' to continue.

UFCG |: Enter the set of genes ... (--set) : PRO

We do not want to maintain temporary files. Type 'n'.

UFCG |: Maintain temporarily created files? (y/n) : n

Now the pipeline will automatically solve dependency and inform you to enter the path of the binaries remaining unsolved.
After then, the program will ask you to decide the number of CPU threads to use. Allocate appropriate amount of threads considering the hardware specification and your affordable time. In here, 10 threads are allocated as an example.

UFCG |: Solving dependencies...

UFCG |: Enter the number ... (--thread, default = 1) : 10

Then the pipeline will ask for the metadata. In this tutorial, we will use the prepared table with complete information.

UFCG |: Provide metadata ... (y/n) : y

UFCG |: Enter the file ... (--metadata) : sample/meta_full.tsv

Finally, after refusing to configurate the advanced options, the pipeline will display the single-line command to run, and ask for your final confirmation. Enter 'y' to launch the pipeline.

UFCG |: Follwing command will be executed : ufcg profile --input sample/seq/ --output sample/out/ --set PRO --thread 10 --metadata sample/meta_full.tsv

UFCG |: Confirm? Enter y (confirm) / n (reset) / x (exit) : y

Tip. If you copy and save the command, you can skip the entire interactive process next time.

Extraction process

By launching the pipeline, the process will begin by reading the input files in and preparing queries to process.
After the preparation finishes, you will see the visualized progress bar such as following.

UFCG |: Reading input data...

UFCG |: Queries prepared. 6 genome sequences identified.

UFCG |: QUERY 1/6 : GCA_000091045.1 (Cryptococcus neoformans)

UFCG |: Extracting protein markers...

UFCG |: PROGRESS : [XXXXXXXXXX...XXX]

Alphabets in the progress bar, which dynamically change during the extraction process, indicate the current status of the genes from the UFCG set as following.

As the extraction finishes, the program will report the result by the count of genes based on their existence, and write the .ucg file on the output folder.

UFCG |: RESULT : [Single: 52 ; Duplicated : 3 ; Missing : 6]

UFCG |: Writing results on : sample/out/GCA_000091045.1.ucg

Check results

The pipeline will automatically terminate as it finishes extracting the gene profiles from all six queries. You can check the result files lying in the output directory by entering the following command.

$ ls -1 sample/out
GCA_000091045.1.ucg
GCA_000146045.2.ucg
GCA_000149205.2.ucg
GCA_000182565.2.ucg
GCA_000271585.1.ucg
GCA_000697725.1.ucg

Infer phylogenetic tree UFCG tree

By following this guide, you will

• infer phylogenetic relationship of the sample species using previously extracted core genes,
• draw their gene trees on sample/tree/tutorial/ directory,
• and learn how to replace the labels of tree with desired information.

Install FastTree

Since we will use FastTree as a phylogenetic tree reconstruction tool, you have to install it first. If you are using conda, enter following command to install FastTree on your environment.

$ conda install -c bioconda fasttree

Otherwise, you can download the binary from here and install it manually.

Run tree module

Launch the module by entering the directory with profiles as an input, labeling them with unique IDs. Since we do not want to perform precise phylogeny here, we will use FastTree.

$ ufcg tree -i sample/out/ -l uid -o sample/tree/ -n tutorial -t 10 -p fasttree

The pipeline will automatically detect .ucg files from sample/out/ directory, perform multiple alignment on each core gene, and reconstruct the phylogenetic tree along with the single gene trees on sample/tree/tutorial/ directory.

Check results

Run following command to check the contents of the result directory.

$ ls -1 sample/tree/tutorial/
align
aligned_concatenated.fasta
concatenated.nwk
concatenated_gsi_50.nwk
misc
tree
tutorial.log
tutorial.trm

• align directory contains FASTA formatted single gene alignments of 68 core genes.
• tree directory contains Newick formatted single gene trees.

Reconstructed phylogenetic tree is written in concatenated.nwk. Printing the Newick tree in a plain text will display its content such as following (unique ID number may differ).

$ cat sample/tree/tutorial/concatenated.nwk
((8255661378167235541:0.339741572,8500290590050038758:0.372848133)
1.000:0.166894959,(653512030666424077:0.477038224,8527107384299568166:0.366508240)
1.000:0.164010396,(4422721812785568721:0.391354384,7201562467530039230:0.367452677)1.000:0.096539693);

Oops! Since we labelled the leaves with their unique IDs, this tree does not inform us the actual taxonomic relationship in a glance. We have to replace their label with proper metadata to solve this problem.

Replace labels with taxonomic information UFCG prune

Launch prune module with following command, indicating it to replace the labels of UFCG tree with taxon names and strain names.

$ ufcg prune -i sample/tree/tutorial/tutorial.trm -g UFCG -l taxon,strain

This process will create the replaced UFCG tree as replaced.UFCG.nwk in current directory. Move it to our tutorial directory and check its content.

$ mv replaced.UFCG.nwk sample/tree/tutorial/replaced.nwk

$ cat sample/tree/tutorial/replaced.nwk
((Cryptococcus_neoformans|JEC21:0.339741572,Trametes_versicolor|FP-101664_SS1:0.372848133)
1.000:0.166894959,(Saccharomyces_cerevisiae|S288C:0.477038224,Aspergillus_nidulans|FGSC_A4:0.366508240)
1.000:0.164010396,(Spizellomyces_punctatus|DAOM_BR117:0.391354384,Rhizopus_oryzae|99-892:0.367452677)
1.000:0.096539693);

Visualize result with MEGA 11

MEGA, provided by MEGA software, is used as a visualization toolkit in this tutorial. Use this external link to download the software.

Launch the MEGA application and load up the replaced tree with Analysis - User Tree - Display Newick Trees menu. The tree will be visualized such as following image.

By rearranging the subtrees and labeling them with phyla, the tree can be modified like following.

Considering Spizellomyces punctatus as a root, you can easily see that the classification of remaining species are intact, corresponding to the known phylogenetic relationship of fungal phyla.