Skip to content

Instantly share code, notes, and snippets.

View neksa's full-sized avatar

Alexander Goncearenco neksa

41 followers · 33 following
View GitHub Profile
@neksa
neksa / MDS.py
Created May 11, 2016 16:43
import numpy as np
import matplotlib.pyplot as plt
from sklearn import manifold
%matplotlib inline
D = np.array([[0, 10, 6], [10, 0, 5], [6, 5, 0]])
M = manifold.MDS(n_components=2, n_init=1, max_iter=10000, metric=True, dissimilarity="precomputed")
K = M.fit_transform(D)
print("Stress", M.stress_)
@neksa
neksa / Mahalanobis_Outliers.ipynb
Created June 8, 2016 21:43 — forked from kevindavenport/Mahalanobis_Outliers.ipynb
An IPython notebook created for my blog post on http://kldavenport.com about the Mahalanobis Distance function and outlier detection.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
@neksa
neksa / chr_rs_acc_via_eutils.py
Created April 25, 2017 20:53 — forked from eweitz/chr_rs_acc_via_eutils.py
NCBI EUtils demo: get a chromosome's RefSeq accession given its name and assembly
'''
This simple script shows how to use NCBI E-Utilies to get a chromosome's
RefSeq accession given the chromosome's name and its genome assembly.
Example:
$ python3 chr_rs_acc_via_eutils.py
RefSeq accession for chromosome 6 on genome assembly GRCh38 (GCF_000001405.26):
NC_000006.12
'''
@neksa
neksa / gist:4c8acf8b09a2db6c1cf4436cd0c9bbd9
Created October 9, 2017 21:32
# test the new generatd machine learning algorithm
def testAlgorithm(algorithm, trainSet, trainSetAnswers):
print("Type of trainSetAnswers is " + str(type(trainSetAnswers)))
for i in range(len(trainSetAnswers)):
y_predicted = algorithm.predict(trainSet[index])
correctAnswer = trainSetAnswers[index]
print("Correct Answer is " + str(correctAnswer))
correctAnswer = correctAnswer.reshape(1,-1)
print("Shape of CorrectAnswer is " + str(correctAnswer.shape))
@neksa
neksa / test_api.py
Created January 20, 2018 01:40
import requests
import json
MUTAGENE_URL = "https://www.ncbi.nlm.nih.gov/research/mutagene"
def get_profile(fname, assembly=37):
"""
Calling MutaGene REST API to convert a VCF file into a mutational profile (96 context-dependent mutational probabilities)
and profile_counts (counts of mutations for each of the 96 context-dependent mutations)
@neksa
neksa / TP53.diff
Created May 3, 2018 19:05
3a4
> M1L 0.539601 0 0
5a7
> M1V 1.030088 0 0
172,175c174,177
< L25L 3.182415 0 0
< L25P 0.907837 0 0
< L25Q 0.440429 0 0
< L25R 0.202944 0 0
---
@neksa
neksa / asn1.py
Created May 30, 2018 21:39
import zlib
from pyasn1.codec.ber import decoder
def decode_residues(res):
residues = []
try:
code, rest = decoder.decode(zlib.decompress(res, 16 + zlib.MAX_WBITS))
except:
pass
for i in range(len(code)):
@neksa
neksa / mutation_context.py
Created June 26, 2018 18:11
try:
import twobitreader as tbr
except:
print("twobitreader module required")
nucleotides = "ACGT" # this order of nucleotides is important for reversing
complementary_nucleotide = dict(zip(nucleotides, reversed(nucleotides)))
TWOBIT_GENOMES_PATH = '/net/pan1/mutagene/data/genomes/'
@neksa
neksa / test_motifs_api.py
Last active August 8, 2018 13:36
import requests
import json
# MUTAGENE_URL = "https://www.ncbi.nlm.nih.gov/research/mutagene"
# MUTAGENE_URL = "https://dev.ncbi.nlm.nih.gov/research/mutagene"
MUTAGENE_URL = "https://mwebdev2/research/mutagene"
# MUTAGENE_URL = "http://localhost:5000"
def get_motifs(fname, assembly=37):
@neksa
neksa / scatterplot_matrix.py
Last active August 8, 2018 18:45
%matplotlib inline
import seaborn as sns
sns.set(style="ticks")
df = sns.load_dataset("iris")
sns.pairplot(df, x_vars=["sepal_length", "sepal_width"], y_vars=["petal_length", "petal_width"], markers=".", kind="reg")
# round(df.corr()**2, 2)
sns.heatmap(df.corr(), cmap=sns.color_palette("RdBu_r", 7), annot=True, fmt=".2f", center=0, vmin=-1.0, vmax=1.0)