-
Notifications
You must be signed in to change notification settings - Fork 0
Tutorial
The standard input format of StructMAn is called simple mutation list format (.smlf). It is a tab or space separated text file format containing the proteomic positions to be mapped and analyzed by StructMAn.
Examples would look like:
P04637 annotate_whole_protein
P04637-3 annotate_isoform
Q96HA4 A31 annotate_specific_position
Q96HA4 P50 annotate_another_specific_position
Q96HA4 77 annotate_position_without_giving_amino_acid_type_also_possible
Q96HA4 P200V annotate_SNV
Q96HA4 P200A annotate_another_SNV
Q96HA4 200D also_possible
Q16637 P210_P241del deletion_example
P04439 S337_D338insGGEGVK insertion_example
P04637 R333_R337delinsKGKEK substitution_example
1JOO:A PDB_style_input_annotates whole chain
1DPO:A M180 specific_position_PDB_style
1DPO:A V181_A SNVs_need_an_underscore_in_PDB_style
1DPO:A K188A this_annotates_position_188_with_insertion_code_A
1DPO:A K188A_A this_codes_the_SNV_of_lysine_position_188_with_insertion_code_A_to_alanine
1DPO:A G188_A this_codes_the_SNV_of_glycine_position_188_to_alanine
StructMAn can also process whole proteins given by a file in fasta-format.
StructMAn is a fully automated pipeline. Only one command is needed in order to process one input dataset placed in the designated input folder:
structman
or, if the input file is not in that folder:
structman -i [path to the input file]
If you want to perform a simple run that would take only one line of a .smlf-file as input, there is no need for creating a separate input file. Instead of giving a path to an input file, simply give the input with the '-i' flag:
structman -i Q9Q288
There are multiple levels of verbosity available by adding --verbosity <0-5> to the command (default is level 1).
The results are written into the designated output folder. There are two default output files:
- the classification table
- the statistics file
- the feature table
StructMAn is able to maintain once processed results in a local database. More info about this feature can be found here
Most results provided by StructMAn are derived from structural information gained from the PDB. Hence a local instance of it can significantly reduce the computation times of StructMAn. Here is explained how to create local resources.