23 lines (21 with data), 12.4 kB

ÇĽE0îdocutils.nodesöîdocumentöôö)üö}ö(î	rawsourceöîöîchildrenö]öhîsectionöôö)üö}ö(hhh]ö(hîtitleöôö)üö}ö(hî*Multi-omics Autoencoder Integration (maui)öh]öhîTextöôöî*Multi-omics Autoencoder Integration (maui)öůöüö}ö(hhîparentöhh
hîsourceöNîlineöNubaî
attributesö}ö(îidsö]öîclassesö]öînamesö]öîdupnamesö]öîbackrefsö]öuîtagnameöhhhh
hhî+/home/jona/work/phd/maui/maui/doc/index.rstöhKubhî	paragraphöôö)üö}ö(hî§maui is an autoencoder-based framework for multi-omics data analysis. It consists of two main modules, :doc:`maui`, and :doc:`utils`. For an introduction of the use of autoencoders for multi-omics integration, see :doc:`autoencoder-integration`.öh]ö(hîgmaui is an autoencoder-based framework for multi-omics data analysis. It consists of two main modules, öůöüö}ö(hîgmaui is an autoencoder-based framework for multi-omics data analysis. It consists of two main modules, öhh-h
hhNhNubîsphinx.addnodesöîpending_xreföôö)üö}ö(hî:doc:`maui`öh]öhîinlineöôö)üö}ö(hh;h]öhîmauiöůöüö}ö(hhhh?ubah}ö(h]öh!]ö(îxreföîstdöîstd-docöeh#]öh%]öh']öuh)h=hh9ubah}ö(h]öh!]öh#]öh%]öh']öîreftypeöîdocöî	refdomainöhJîrefexplicitöëî	reftargetöîmauiöîrefdocöîindexöîrefwarnöłuh)h7hh*hKhh-ubhî, and öůöüö}ö(hî, and öhh-h
hhNhNubh8)üö}ö(hî:doc:`utils`öh]öh>)üö}ö(hheh]öhîutilsöůöüö}ö(hhhhgubah}ö(h]öh!]ö(hIîstdöîstd-docöeh#]öh%]öh']öuh)h=hhcubah}ö(h]öh!]öh#]öh%]öh']öîreftypeöîdocöî	refdomainöhqîrefexplicitöëhYîutilsöh[h\h]łuh)h7hh*hKhh-ubhîR. For an introduction of the use of autoencoders for multi-omics integration, see öůöüö}ö(hîR. For an introduction of the use of autoencoders for multi-omics integration, see öhh-h
hhNhNubh8)üö}ö(hî:doc:`autoencoder-integration`öh]öh>)üö}ö(hhłh]öhîautoencoder-integrationöůöüö}ö(hhhhŐubah}ö(h]öh!]ö(hIîstdöîstd-docöeh#]öh%]öh']öuh)h=hhćubah}ö(h]öh!]öh#]öh%]öh']öîreftypeöîdocöî	refdomainöhöîrefexplicitöëhYîautoencoder-integrationöh[h\h]łuh)h7hh*hKhh-ubhî
.öůöüö}ö(hî
.öhh-h
hhNhNubeh}ö(h]öh!]öh#]öh%]öh']öuh)h+hh*hKhhh
hubh
)üö}ö(hhh]ö(h)üö}ö(hîTable of contentsöh]öhîTable of contentsöůöüö}ö(hh┤hh▓h
hhNhNubah}ö(h]öh!]öh#]öh%]öh']öuh)hhh»h
hhh*hKubhîcompoundöôö)üö}ö(hhh]öh6îtoctreeöôö)üö}ö(hhh]öh}ö(h]öh!]öh#]öh%]öh']öhh\îentriesö]ö(Nîautoencoder-integrationöćöNîdata-normalizationöćöNî$filtering-and-merging-latent-factorsöćöNîsaving-and-loading-modelsöćöNîmauiöćöNîutilsöćöeîincludefilesö]ö(hĎhďhÍhěh┌h▄eîmaxdepthöKîcaptionöNîgloböëîhiddenöëî
includehiddenöëînumberedöKî
titlesonlyöëuh)h┼hh*hK
hh┬ubah}ö(h]öh!]öîtoctree-wrapperöah#]öh%]öh']öuh)h└hh»h
hhh*hNubeh}ö(h]öîtable-of-contentsöah!]öh#]öîtable of contentsöah%]öh']öuh)h	hhh
hhh*hKubh
)üö}ö(hhh]ö(h)üö}ö(hî
Quickstartöh]öhî
Quickstartöůöüö}ö(hhűhh¨h
hhNhNubah}ö(h]öh!]öh#]öh%]öh']öuh)hhh÷h
hhh*hKubh,)üö}ö(hîÎThe ``Maui`` class implements ``scikit-learn``'s ``BaseEstimator``. In order to infer latent factors in multi-omics data, first instantiate a ``Maui`` model with the desired parameters, and then fit it to some data:öh]ö(hîThe öůöüö}ö(hîThe öhj
h
hhNhNubhîliteralöôö)üö}ö(hî``Maui``öh]öhîMauiöůöüö}ö(hhhj
ubah}ö(h]öh!]öh#]öh%]öh']öuh)j
hj
ubhî class implements öůöüö}ö(hî class implements öhj
h
hhNhNubj
)üö}ö(hî``scikit-learn``öh]öhîscikit-learnöůöüö}ö(hhhj%
ubah}ö(h]öh!]öh#]öh%]öh']öuh)j
hj
ubhîÔÇÖs öůöüö}ö(hî's öhj
h
hhNhNubj
)üö}ö(hî``BaseEstimator``öh]öhî
BaseEstimatoröůöüö}ö(hhhj8
ubah}ö(h]öh!]öh#]öh%]öh']öuh)j
hj
ubhîL. In order to infer latent factors in multi-omics data, first instantiate a öůöüö}ö(hîL. In order to infer latent factors in multi-omics data, first instantiate a öhj
h
hhNhNubj
)üö}ö(hî``Maui``öh]öhîMauiöůöüö}ö(hhhjK
ubah}ö(h]öh!]öh#]öh%]öh']öuh)j
hj
ubhîA model with the desired parameters, and then fit it to some data:öůöüö}ö(hîA model with the desired parameters, and then fit it to some data:öhj
h
hhNhNubeh}ö(h]öh!]öh#]öh%]öh']öuh)h+hh*hKhh÷h
hubhî
literal_blocköôö)üö}ö(hîáfrom maui import Maui

maui_model = maui.Maui(n_hidden=[900], n_latent=70, epochs=100)
z = maui_model.fit_transform({'mRNA': gex, 'Mutations': mut, 'CNV': cnv})öh]öhîáfrom maui import Maui

maui_model = maui.Maui(n_hidden=[900], n_latent=70, epochs=100)
z = maui_model.fit_transform({'mRNA': gex, 'Mutations': mut, 'CNV': cnv})öůöüö}ö(hhhjf
ubah}ö(h]öh!]öh#]öh%]öh']öî	xml:spaceöîpreserveöîlanguageöîpythonöîlinenosöëîhighlight_argsö}öuh)jd
hh*hKhh÷h
hubh,)üö}ö(hX┴
This will instantiate a maui model with one hidden layer of 900 nodes, and a middle layer of 70 nodes, which will be traiend for 100 epochs. It then feeds the multi-omics data in ``gex``, ``mut``, and ``cnv`` to the fitting procedure. The omics data (``gex`` et. al.) are ``pandas.DataFrame`` objects of dimension (n_features, n_samples). The return object ``z`` is a ``pandas.DataFrame`` (n_samples, n_latent), and may be used for further analysis.öh]ö(hî│This will instantiate a maui model with one hidden layer of 900 nodes, and a middle layer of 70 nodes, which will be traiend for 100 epochs. It then feeds the multi-omics data in öůöüö}ö(hî│This will instantiate a maui model with one hidden layer of 900 nodes, and a middle layer of 70 nodes, which will be traiend for 100 epochs. It then feeds the multi-omics data in öhj{
h
hhNhNubj
)üö}ö(hî``gex``öh]öhîgexöůöüö}ö(hhhjä
ubah}ö(h]öh!]öh#]öh%]öh']öuh)j
hj{
ubhî, öůöüö}ö(hî, öhj{
h
hhNhNubj
)üö}ö(hî``mut``öh]öhîmutöůöüö}ö(hhhjŚ
ubah}ö(h]öh!]öh#]öh%]öh']öuh)j
hj{
ubhî, and öůöüö}ö(hî, and öhj{
h
hhNhNubj
)üö}ö(hî``cnv``öh]öhîcnvöůöüö}ö(hhhj¬
ubah}ö(h]öh!]öh#]öh%]öh']öuh)j
hj{
ubhî+ to the fitting procedure. The omics data (öůöüö}ö(hî+ to the fitting procedure. The omics data (öhj{
h
hhNhNubj
)üö}ö(hî``gex``öh]öhîgexöůöüö}ö(hhhjŻ
ubah}ö(h]öh!]öh#]öh%]öh']öuh)j
hj{
ubhî et. al.) are öůöüö}ö(hî et. al.) are öhj{
h
hhNhNubj
)üö}ö(hî``pandas.DataFrame``öh]öhîpandas.DataFrameöůöüö}ö(hhhjđ
ubah}ö(h]öh!]öh#]öh%]öh']öuh)j
hj{
ubhîA objects of dimension (n_features, n_samples). The return object öůöüö}ö(hîA objects of dimension (n_features, n_samples). The return object öhj{
h
hhNhNubj
)üö}ö(hî``z``öh]öhî
zöůöüö}ö(hhhjŃ
ubah}ö(h]öh!]öh#]öh%]öh']öuh)j
hj{
ubhî is a öůöüö}ö(hî is a öhj{
h
hhNhNubj
)üö}ö(hî``pandas.DataFrame``öh]öhîpandas.DataFrameöůöüö}ö(hhhj÷
ubah}ö(h]öh!]öh#]öh%]öh']öuh)j
hj{
ubhî= (n_samples, n_latent), and may be used for further analysis.öůöüö}ö(hî= (n_samples, n_latent), and may be used for further analysis.öhj{
h
hhNhNubeh}ö(h]öh!]öh#]öh%]öh']öuh)h+hh*hK"hh÷h
hubh,)üö}ö(hî]In order to check the model's convergance, the ``hist`` object may be inspected, and plotted:öh]ö(hî1In order to check the modelÔÇÖs convergance, the öůöüö}ö(hî/In order to check the model's convergance, the öhjh
hhNhNubj
)üö}ö(hî``hist``öh]öhîhistöůöüö}ö(hhhjubah}ö(h]öh!]öh#]öh%]öh']öuh)j
hjubhî& object may be inspected, and plotted:öůöüö}ö(hî& object may be inspected, and plotted:öhjh
hhNhNubeh}ö(h]öh!]öh#]öh%]öh']öuh)h+hh*hK$hh÷h
hubje
)üö}ö(hîmaui_model.hist.plot()öh]öhîmaui_model.hist.plot()öůöüö}ö(hhhj1ubah}ö(h]öh!]öh#]öh%]öh']öjt
ju
jv
îpythonöjx
ëjy
}öuh)jd
hh*hK&hh÷h
hubhîimageöôö)üö}ö(hî.. image:: _static/hist.png
öh]öh}ö(h]öh!]öh#]öh%]öh']öîuriöî_static/hist.pngöî
candidatesö}öî
*öjNsuh)jAhh÷h
hhh*hK+ubh,)üö}ö(hîîFor a more comprehensive example, check out `our vignette <https://github.com/BIMSBbioinfo/maui/blob/master/vignette/maui_vignette.ipynb>`_.öh]ö(hî,For a more comprehensive example, check out öůöüö}ö(hî,For a more comprehensive example, check out öhjRh
hhNhNubhî	referenceöôö)üö}ö(hî_`our vignette <https://github.com/BIMSBbioinfo/maui/blob/master/vignette/maui_vignette.ipynb>`_öh]öhîour vignetteöůöüö}ö(hhhj]ubah}ö(h]öh!]öh#]öh%]öh']öînameöîour vignetteöîrefuriöîMhttps://github.com/BIMSBbioinfo/maui/blob/master/vignette/maui_vignette.ipynböuh)j[hjRubhîtargetöôö)üö}ö(hîP <https://github.com/BIMSBbioinfo/maui/blob/master/vignette/maui_vignette.ipynb>öh]öh}ö(h]öîour-vignetteöah!]öh#]öîour vignetteöah%]öh']öîrefuriöjnuh)joî
referencedöK
hjRubhî
.öůöüö}ö(hhĘhjRh
hhNhNubeh}ö(h]öh!]öh#]öh%]öh']öuh)h+hh*hK,hh÷h
hubh
)üö}ö(hhh]ö(h)üö}ö(hîIndices and tablesöh]öhîIndices and tablesöůöüö}ö(hjÄhjîh
hhNhNubah}ö(h]öh!]öh#]öh%]öh']öuh)hhjëh
hhh*hK0ubhîbullet_listöôö)üö}ö(hhh]ö(hî	list_itemöôö)üö}ö(hî:ref:`genindex`öh]öh,)üö}ö(hjúh]öh8)üö}ö(hjúh]öh>)üö}ö(hjúh]öhîgenindexöůöüö}ö(hhhjźubah}ö(h]öh!]ö(hIîstdöîstd-reföeh#]öh%]öh']öuh)h=hjĘubah}ö(h]öh!]öh#]öh%]öh']öîreftypeöîreföî	refdomainöjÁîrefexplicitöëhYîgenindexöh[h\h]łuh)h7hh*hK2hjąubah}ö(h]öh!]öh#]öh%]öh']öuh)h+hh*hK2hjíubah}ö(h]öh!]öh#]öh%]öh']öuh)jčhjťh
hhh*hNubjá)üö}ö(hî:ref:`modindex`öh]öh,)üö}ö(hjËh]öh8)üö}ö(hjËh]öh>)üö}ö(hjËh]öhîmodindexöůöüö}ö(hhhj█ubah}ö(h]öh!]ö(hIîstdöîstd-reföeh#]öh%]öh']öuh)h=hjěubah}ö(h]öh!]öh#]öh%]öh']öîreftypeöîreföî	refdomainöjňîrefexplicitöëhYîmodindexöh[h\h]łuh)h7hh*hK3hjŇubah}ö(h]öh!]öh#]öh%]öh']öuh)h+hh*hK3hjĐubah}ö(h]öh!]öh#]öh%]öh']öuh)jčhjťh
hhh*hNubjá)üö}ö(hî
:ref:`search`öh]öh,)üö}ö(hjh]öh8)üö}ö(hjh]öh>)üö}ö(hjh]öhîsearchöůöüö}ö(hhhjubah}ö(h]öh!]ö(hIîstdöîstd-reföeh#]öh%]öh']öuh)h=hjubah}ö(h]öh!]öh#]öh%]öh']öîreftypeöîreföî	refdomainöjîrefexplicitöëhYîsearchöh[h\h]łuh)h7hh*hK4hjubah}ö(h]öh!]öh#]öh%]öh']öuh)h+hh*hK4hj
ubah}ö(h]öh!]öh#]öh%]öh']öuh)jčhjťh
hhh*hNubeh}ö(h]öh!]öh#]öh%]öh']öîbulletöjQuh)jÜhh*hK2hjëh
hubeh}ö(h]öîindices-and-tablesöah!]öh#]öîindices and tablesöah%]öh']öuh)h	hh÷h
hhh*hK0ubeh}ö(h]öî
quickstartöah!]öh#]öî
quickstartöah%]öh']öuh)h	hhh
hhh*hKubeh}ö(h]öî(multi-omics-autoencoder-integration-mauiöah!]öh#]öî*multi-omics autoencoder integration (maui)öah%]öh']öuh)h	hhh
hhh*hKubah}ö(h]öh!]öh#]öh%]öh']öîsourceöh*uh)h
îcurrent_sourceöNîcurrent_lineöNîsettingsöîdocutils.frontendöîValuesöôö)üö}ö(hNî	generatoröNî	datestampöNîsource_linköNî
source_urlöNî
toc_backlinksöîentryöîfootnote_backlinksöK
î
sectnum_xformöK
îstrip_commentsöNîstrip_elements_with_classesöNî
strip_classesöNîreport_levelöKî
halt_levelöKîexit_status_levelöKîdebugöNîwarning_streamöNî	tracebacköłîinput_encodingöî	utf-8-sigöîinput_encoding_error_handleröîstrictöîoutput_encodingöîutf-8öîoutput_encoding_error_handleröjsîerror_encodingöîUTF-8öîerror_encoding_error_handleröîbackslashreplaceöî
language_codeöîenöîrecord_dependenciesöNîconfigöNî	id_prefixöhîauto_id_prefixöîidöî
dump_settingsöNîdump_internalsöNîdump_transformsöNîdump_pseudo_xmlöNîexpose_internalsöNîstrict_visitoröNî_disable_configöNî_sourceöh*î_destinationöNî
_config_filesö]öîpep_referencesöNîpep_base_urlöî https://www.python.org/dev/peps/öîpep_file_url_templateöîpep-%04döîrfc_referencesöNîrfc_base_urlöîhttps://tools.ietf.org/html/öî	tab_widthöKîtrim_footnote_reference_spaceöëîfile_insertion_enabledöłîraw_enabledöK
îsyntax_highlightöîlongöîsmart_quotesöłîsmartquotes_localesö]öîcharacter_level_inline_markupöëîdoctitle_xformöëî
docinfo_xformöK
îsectsubtitle_xformöëîembed_stylesheetöëîcloak_email_addressesöłîenvöNubîreporteröNîindirect_targetsö]öîsubstitution_defsö}öîsubstitution_namesö}öîrefnamesö}öîrefidsö}öînameidsö}ö(jMjJhˇh­jEjBjzjwj=j:uî	nametypesö}ö(jMNhˇNjENjzłj=Nuh}ö(jJhh­h»jBh÷jwjqj:jëuî
footnote_refsö}öî
citation_refsö}öî
autofootnotesö]öîautofootnote_refsö]öîsymbol_footnotesö]öîsymbol_footnote_refsö]öî	footnotesö]öî	citationsö]öîautofootnote_startöK
îsymbol_footnote_startöKîid_startöK
îparse_messagesö]öîtransform_messagesö]öîtransformeröNî
decorationöNh
hub.