AttributeError: type object 'DataFrame' has no attribute 'from_csv'

[lijing28101]

Hi, my code is

portcullis full --threads 28 --verbose --use_csi --output portcullis_out --orientation FR Saccharomyces_cerevisiae.fa Saccharomyces_cerevisiae_rnaseq_sorted.bam

When I run to the filter step, it shows the error

xecuting python script with this command: portcullis/rule_filter.py portcullis/rule_filter.py --pos_json //pylon5/mc5pl7p/lijingqb/bin/anaconda3/envs/mikado/share/portcullis/balanced/selftrain_initial_pos.layer1.json //pylon5/mc5pl7p/lijingqb/bin/anaconda3/envs/mikado/share/portcullis/balanced/selftrain_initial_pos.layer2.json //pylon5/mc5pl7p/lijingqb/bin/anaconda3/envs/mikado/share/portcullis/balanced/selftrain_initial_pos.layer3.json --neg_json //pylon5/mc5pl7p/lijingqb/bin/anaconda3/envs/mikado/share/portcullis/balanced/selftrain_initial_neg.layer1.json //pylon5/mc5pl7p/lijingqb/bin/anaconda3/envs/mikado/share/portcullis/balanced/selftrain_initial_neg.layer2.json //pylon5/mc5pl7p/lijingqb/bin/anaconda3/envs/mikado/share/portcullis/balanced/selftrain_initial_neg.layer3.json //pylon5/mc5pl7p/lijingqb/bin/anaconda3/envs/mikado/share/portcullis/balanced/selftrain_initial_neg.layer4.json //pylon5/mc5pl7p/lijingqb/bin/anaconda3/envs/mikado/share/portcullis/balanced/selftrain_initial_neg.layer5.json //pylon5/mc5pl7p/lijingqb/bin/anaconda3/envs/mikado/share/portcullis/balanced/selftrain_initial_neg.layer6.json //pylon5/mc5pl7p/lijingqb/bin/anaconda3/envs/mikado/share/portcullis/balanced/selftrain_initial_neg.layer7.json --prefix=portcullis_out/3-filt/portcullis_filtered.selftrain.initialset portcullis_out/2-junc/portcullis_all.junctions.tab
Loading input junctions ... Traceback (most recent call last):
  File "//pylon5/mc5pl7p/lijingqb/bin/anaconda3/envs/mikado/share/portcullis/scripts/portcullis/rule_filter.py", line 392, in <module>
    if __name__=='__main__': main()
  File "//pylon5/mc5pl7p/lijingqb/bin/anaconda3/envs/mikado/share/portcullis/scripts/portcullis/rule_filter.py", line 385, in main
    create_training_sets(args)
  File "//pylon5/mc5pl7p/lijingqb/bin/anaconda3/envs/mikado/share/portcullis/scripts/portcullis/rule_filter.py", line 127, in create_training_sets
    original = DataFrame.from_csv(args.input, sep='\t', header=0)
AttributeError: type object 'DataFrame' has no attribute 'from_csv'

Portcullis completed.
Total runtime: 69.6s

../lib/include/portcullis/python_helper.hpp(167): Throw in function void portcullis::PyHelper::execute(std::__cxx11::string, int, char**)
Dynamic exception type: boost::exception_detail::clone_impl<portcullis::PortcullisPythonException>
std::exception::what: std::exception

I found that DataFrame doesn't have from_csv any more, it should be pd.read_csv.
I manually revised the code in rule_filter.py. However, it gave me a new error

src/junction.cc(1166): Throw in function static std::shared_ptr<portcullis::Junction> portcullis::Junction::parse(const string&)
Dynamic exception type: boost::exception_detail::clone_impl<portcullis::JunctionException>
std::exception::what: std::exception
[portcullis::JunctionError*] = Could not parse line due to incorrect number of columns.  This is probably a version mismatch.  Check file and portcullis versions.  Expected 75 columns.  Found 76.  Line: 37	37	0	I	230218	87387	87499	113	87312	87574	+	+	+	GT	AG	C	0	0	0	2191	541	2191	0	2191	0	2169	2142149	0.9808309999999999	0	1503	688	0	4.88665	0.0826107	75	38	38	0	10	5	0	4808	10375	4808	0	0	0	0	1	2185	2161	2123	2083	2035	1993	1969	1953	1944	1901864	1852	1810	1790	1749	1396	1361	1318	1082	871

I use pyhton 3.6
Please help me to solve the problem.
Thanks!

[lucventurini]

Dear @lijing28101 , it appears you are using an older version of Portcullis - we solved this bug about half a year ago. How did you install the program?

In case you are using BioConda to install it, we did update Portcullis there. A conda update -c bioconda portcullis should install the latest, fixed version.

Kind regards

Luca Venturini

[lijing28101]

Hi @lucventurini , I cannot install version 1.2.2. through conda. It shows package conflict. But my environment just install mikado. I think their dependency should be consistent.

Package openssl conflicts for:
defaults::python[version='>=3.6'] -> openssl[version='1.0.*|1.0.*,>=1.0.2l,<1.0.3a|>=1.0.2m,<1.0.3a|>=1.0.2n,<1.0.3a|>=1.0.2o,<1.0.3a|>=1.0.2p,<1.0.3a|>=1.1.1a,<1.1.2a|>=1.1.1b,<1.1.2a|>=1.1.1c,<1.1.2a|>=1.1.1d,<1.1.2a']
Package samtools conflicts for:
portcullis=1.2.2 -> samtools[version='>=1.9']
Package tk conflicts for:
defaults::python[version='>=3.6'] -> tk[version='8.6.*|>=8.6.7,<8.7.0a0|>=8.6.8,<8.7.0a0']
Package readline conflicts for:
defaults::python[version='>=3.6'] -> readline[version='7.*|>=7.0,<8.0a0']
Package libstdcxx-ng conflicts for:
portcullis=1.2.2 -> libstdcxx-ng[version='>=7.3.0']
defaults::python[version='>=3.6'] -> libstdcxx-ng[version='>=7.2.0|>=7.3.0']
Package libgcc-ng conflicts for:
portcullis=1.2.2 -> libgcc-ng[version='>=7.3.0']
defaults::python[version='>=3.6'] -> libgcc-ng[version='>=7.2.0|>=7.3.0']
Package boost-cpp conflicts for:
portcullis=1.2.2 -> boost-cpp[version='>=1.70.0,<1.70.1.0a0']
Package ncurses conflicts for:
defaults::python[version='>=3.6'] -> ncurses[version='6.0.*|>=6.0,<7.0a0|>=6.1,<7.0a0']
Package pandas conflicts for:
portcullis=1.2.2 -> pandas
Package pip conflicts for:
defaults::python[version='>=3.6'] -> pip
Package zlib conflicts for:
portcullis=1.2.2 -> zlib[version='>=1.2.11,<1.3.0a0']
defaults::python[version='>=3.6'] -> zlib[version='>=1.2.11,<1.3.0a0']
Package tabulate conflicts for:
portcullis=1.2.2 -> tabulate
Package sqlite conflicts for:
defaults::python[version='>=3.6'] -> sqlite[version='>=3.20.1,<4.0a0|>=3.22.0,<4.0a0|>=3.23.1,<4.0a0|>=3.24.0,<4.0a0|>=3.25.2,<4.0a0|>=3.25.3,<4.0a0|>=3.26.0,<4.0a0|>=3.27.2,<4.0a0|>=3.29.0,<4.0a0|>=3.30.0,<4.0a0|>=3.30.1,<4.0a0']
Package libffi conflicts for:
defaults::python[version='>=3.6'] -> libffi[version='3.2.*|>=3.2.1,<4.0a0']
Package numpy conflicts for:
portcullis=1.2.2 -> numpy
Package xz conflicts for:
defaults::python[version='>=3.6'] -> xz[version='>=5.2.3,<6.0a0|>=5.2.4,<6.0a0']

Best,
Jing

[lucventurini]

Dear @lijing28101,
sorry to hear that. I cannot replicate the problem on my end, however, as the following functioned:

$ conda env remove -n mikado2  # Remove mikado environment if present
$ git clone https://github.com/EI-CoreBioinformatics/mikado.git
$ cd mikado
$ conda env create -f environment.yml  # Create Mikado2 environment
$ conda activate mikado2
$ conda install -c bioconda portcullis  # Showing the log
Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source.
Collecting package metadata (repodata.json): done
Solving environment: done


==> WARNING: A newer version of conda exists. <==
  current version: 4.7.12
  latest version: 4.8.1

Please update conda by running

    $ conda update -n base -c defaults conda



## Package Plan ##

  environment location: /software/stable/python/.conda/x86_64/envs/mikado2

  added / updated specs:
    - portcullis


The following packages will be downloaded:

    package                    |            build
    ---------------------------|-----------------
    certifi-2019.11.28         |           py37_0         153 KB
    htslib-1.10.2              |       h78d89cc_0         1.7 MB  bioconda
    portcullis-1.1.2           |   py37hdbcaa40_0        21.5 MB  bioconda
    samtools-1.10              |       h9402c20_2         343 KB  bioconda
    ------------------------------------------------------------
                                           Total:        23.6 MB

The following NEW packages will be INSTALLED:

  htslib             bioconda/linux-64::htslib-1.10.2-h78d89cc_0
  portcullis         bioconda/linux-64::portcullis-1.1.2-py37hdbcaa40_0
  samtools           bioconda/linux-64::samtools-1.10-h9402c20_2

The following packages will be UPDATED:

  openssl            conda-forge::openssl-1.1.1d-h516909a_0 --> pkgs/main::openssl-1.1.1d-h7b6447c_3

The following packages will be SUPERSEDED by a higher-priority channel:

  certifi                                       conda-forge --> pkgs/main


Proceed ([y]/n)? y


Downloading and Extracting Packages
samtools-1.10        | 343 KB    | ############################################################################################################################################# | 100% 
portcullis-1.1.2     | 21.5 MB   | ############################################################################################################################################# | 100% 
htslib-1.10.2        | 1.7 MB    | ############################################################################################################################################# | 100% 
certifi-2019.11.28   | 153 KB    | ############################################################################################################################################# | 100% 
Preparing transaction: done
Verifying transaction: done
Executing transaction: done
$ pip install dist/Mikado-2.0rc6-cp37-cp37m-linux_x86_64.whl
$ mikado --help
$ portcullis --help

I hope this helps in solving the installation problem on your end.

Kind regards

Luca Venturini

[lijing28101]

Hi @lucventurini , this method install the version 1.1.2, not 1.2.2. And this version has the DataFrame error. I've tried install 1.2.2 from source code, but it also failed when I run make -j 2

[urmi-21]

@lucventurini I deleted my previous comment as @lijing28101 already pointed out that from_csv() is now deprecated.

[lucventurini]

Dear @lijing28101 , @urmi-21 , thank you for your feedback. I am currently revising the Mikado environment definition file to allow installation of both programs in the same conda environment.

Please bear with me while I correct it.

Kind regards

Luca Venturini

[lucventurini]

Dear @lijing28101 , @urmi-21 , the following workaround functions for now: add

  - bioconda::portcullis>=1.2.2

on the dependency list of the environment.yml file I indicated earlier, then create the environment.
This will force conda to install the right versions of the packages

I think that the problem is the fact the portcullis package was created six months ago and therefore its package versions were frozen to earlier versions of zlib. This is quite weird, as the recipe itself does not indicate a specific version for any requirement. I will try to solve the problem on the channel today.

Thank you again for reporting the problem.

[lucventurini]

Update: the problem stems from both zlib and boost having been updated, while the portcullis package requires the earlier version.

[francicco]

Hi,

I'm having the same issue. Did you menage to fix it?

Cheers
F