Class representing an Impute2File.

This is used to either generate a dosage matrix where columns represent variants and rows represent samples or to read the file line by line using the generator syntax.

This also implements the context manager interface.

Usage:

# Read as probabilities (Line tuples).
with open(Impute2File(fn)) as f:
    for line in f:
        # line has name, chrom, pos, a1, a2, probabilities
        print(line)

# Read as dosage.
with open(Impute2File(fn), "dosage") as f:
    for dosage_vector, info in f:
        pass

# Read as a matrix.
with open(Impute2File(fn)) as f:
    # 1 row per sample and 1 column per variant. Values between 0 and 2
    m = f.as_matrix()

If you use the dosage mode, you can also add additional arguments:

• prob_threshold: Genotype probability cutoff for no call values (NaN).

• is_chr23: Not implemented yet, but dosage is computed differently

  for sexual chromosomes for men (hemizygote).

• sex_vector: Not implemented yet, but this is a vector representing

  the gender of every sample (for dosage computation on sexual chromosomes).

as_matrix()[source]¶

Creates a numpy dosage matrix from this file.

Returns:	A numpy matrix where columns represent variant dosage between 0 and 2 and a dataframe describing the variants (major, minor, maf).
Type:	tuple

Warning

This will attempt to load the whole file in memory.

readline()[source]¶

Read a single line from the Impute2File.

This will return either a Line including the genotype probabilities or a dosage vector. This depends on the mode (the second argument given to the file when it was opened).

Available modes are dosage and line.

Parses the SeqXML format representing sequence data.

The returned object will have a list of entries which are Sequence objects.

get_seq(uid)[source]¶

Get a sequence from it’s unique identifier.

Parameters:	uid (str) – The sequence id.