Data Structure
Package for parsing netMHC predictions. Data is stored in a
General Info:
class PredictionCollection
PredictionCollection.dictionary_collections
[ { protein_1 : {'High Affinity Regions : [region1, region 2, ...],
'ProtSeq : 'SNFEDNSKDAJSNFWIAIAIJFNWKSAM....',
'Predictions : [ Prediction,
Prediction,
...
Prediction.Swaps.swap
[ { swap : Prediction },
{ swap : Prediction },
...,
{ swap : Prediction }
]
],
{ protein_3 : {'High Affinity Regions : [region1, region 2, ...],
'ProtSeq : 'SNFEDNSKDAJSNFWIAIAIJFNWKSAM....',
'Predictions : [ Prediction,
Prediction,
...
Prediction.Swaps.swap
[ { swap : Prediction },
{ swap : Prediction },
...,
{ swap : Prediction }
]
],
...,
{ protein_1 : {'High Affinity Regions : [region1, region 2, ...],
'ProtSeq : 'SNFEDNSKDAJSNFWIAIAIJFNWKSAM....',
'Predictions : [ Prediction,
Prediction,
...
Prediction.Swaps.swap
[ { swap : Prediction },
{ swap : Prediction },
...,
{ swap : Prediction }
]
], `
More specific information about the Prediction class:
class Prediction #Holds all the info about a single prediction.
#I.e.:
Prediction.allele
Prediction.original_position
Prediction.protein
Prediction.affinity_level
Prediction.rank
Prediction.core_pep
Prediction.h_avg_ranks
Prediction.n_binders
Prediction.allele
Prediction.nmer`
And as shown above, it also holds data about the possible swaps. For high-affinity peptides (user-defined thershold), the Swap class has as attribute a dictionary with the following structure:
Prediction.Swap.swaps
{'swap1': Prediction,
'swap2': Prediction,
'swap3': Prediction,
...
'swapN': Prediction}
Where swapK is a singly-modified peptide from the original, high-affinity peptide.