search

BIMSBbioinfo / janggu

Deep learning infrastructure for genomics
GNU General Public License v3.0
254 stars 33 forks source link

Classify_nucleotide_sequences #15

Closed eallen4040 closed 4 years ago

eallen4040 commented 4 years ago

Hello!

I am relatively new to the world of neural networks! I've been in contact with this group before and you all have been a tremendous help! I am currently attempting to run the examples available in jupyter notebook. I specifically want to ask about Classify_nucleotide_sequences. I have a couple errors return when I run the first cell:

ImportError Traceback (most recent call last)

in 9 from keras.layers import Maximum 10 ---> 11 from janggu import Janggu 12 from janggu import Scorer 13 from janggu import inputlayer ~/anaconda3/envs/janggu/lib/python3.6/site-packages/janggu/__init__.py in 4 from janggu.decorators import outputconv # noqa 5 from janggu.decorators import outputdense # noqa ----> 6 from janggu.evaluation import Scorer # noqa 7 from janggu.layers import Complement # noqa 8 from janggu.layers import DnaConv2D # noqa ~/anaconda3/envs/janggu/lib/python3.6/site-packages/janggu/evaluation.py in 18 from sklearn.metrics import roc_curve 19 ---> 20 from janggu.utils import ExportJson 21 from janggu.utils import ExportScorePlot 22 from janggu.utils import ExportTsv ~/anaconda3/envs/janggu/lib/python3.6/site-packages/janggu/utils.py in 10 import pandas as pd 11 from Bio import SeqIO ---> 12 from Bio.Alphabet import IUPAC 13 from Bio.Seq import Seq 14 from Bio.SeqRecord import SeqRecord ~/anaconda3/envs/janggu/lib/python3.6/site-packages/Bio/Alphabet/__init__.py in 19 20 raise ImportError( ---> 21 "Bio.Alphabet has been removed from Biopython. In many cases, the alphabet can simply be ignored and removed from scripts. In a few cases, you may need to specify the ``molecule_type`` as an annotation on a SeqRecord for your script to work correctly. Please see https://biopython.org/wiki/Alphabet for more information." 22 ) ImportError: Bio.Alphabet has been removed from Biopython. In many cases, the alphabet can simply be ignored and removed from scripts. In a few cases, you may need to specify the ``molecule_type`` as an annotation on a SeqRecord for your script to work correctly. Please see https://biopython.org/wiki/Alphabet for more information. Can I get a little insight on these errors and how to fix them? I would really appreciate it!
wkopp commented 4 years ago

Hi @eallen4040,

The issue use caused by an update in the package Biopython==1.78. Install Biopython==1.77 with pip install biopython==1.77. This should resolve the issue.

I've adjusted setup.py such that it requires Biopython==1.77. This is available in the github master branch and will be part of the new version soon.

Best, Wolfgang