Using polygenic scores and clinical data for bipolar disorder patient stratification and lithium response prediction: machine learning approach

Micah Cearns, Azmeraw T. Amare, Klaus Oliver Schubert, Anbupalam Thalamuthu, Joseph Frank, Fabian Streit, Mazda Adli, Nirmala Akula, Kazufumi Akiyama, Raffaella Ardau, Bárbara Arias, Jean Michel Aubry, Lena Backlund, Abesh Kumar Bhattacharjee, Frank Bellivier, Antonio Benabarre, Susanne Bengesser, Joanna M. Biernacka, Armin Birner, Clara Brichant-PetitjeanPablo Cervantes, Hsi Chung Chen, Caterina Chillotti, Sven Cichon, Cristiana Cruceanu, Piotr M. Czerski, Nina Dalkner, Alexandre Dayer, Franziska Degenhardt, Maria Del Zompo, J. Raymond Depaulo, Bruno Étain, Peter Falkai, Andreas J. Forstner, Louise Frisen, Mark A. Frye, Janice M. Fullerton, Sébastien Gard, Julie S. Garnham, Fernando S. Goes, Maria Grigoroiu-Serbanescu, Paul Grof, Ryota Hashimoto, Joanna Hauser, Urs Heilbronner, Stefan Herms, Per Hoffmann, Andrea Hofmann, Liping Hou, Yi Hsiang Hsu, Stephane Jamain, Esther Jiménez, Jean Pierre Kahn, Layla Kassem, Po Hsiu Kuo, Tadafumi Kato, John Kelsoe, Sarah Kittel-Schneider, Sebastian Kliwicki, Barbara König, Ichiro Kusumi, Gonzalo Laje, Mikael Landén, Catharina Lavebratt, Marion Leboyer, Susan G. Leckband, Mario Maj, Mirko Manchia, Lina Martinsson, Michael J. McCarthy, Susan McElroy, Francesc Colom, Marina Mitjans, Francis M. Mondimore, Palmiero Monteleone, Caroline M. Nievergelt, Markus M. Nöthen, Tomas Novák, Claire O'Donovan, Norio Ozaki, Vincent Millischer, Sergi Papiol, Andrea Pfennig, Claudia Pisanu, James B. Potash, Andreas Reif, Eva Reininghaus, Guy A. Rouleau, Janusz K. Rybakowski, Martin Schalling, Peter R. Schofield, Barbara W. Schweizer, Giovanni Severino, Tatyana Shekhtman, Paul D. Shilling, Katzutaka Shimoda, Christian Simhandl, Claire M. Slaney, Alessio Squassina, Thomas Stamm, Pavla Stopkova, Fasil Tekola-Ayele, Alfonso Tortorella, Gustavo Turecki, Julia Veeh, Eduard Vieta, Stephanie H. Witt, Gloria Roberts, Peter P. Zandi, Martin Alda, Michael Bauer, Francis J. McMahon, Philip B. Mitchell, Thomas G. Schulze, Marcella Rietschel, Scott R. Clark, Bernhard T. Baune

Research output: Contribution to journalArticlepeer-review


Background Response to lithium in patients with bipolar disorder is associated with clinical and transdiagnostic genetic factors. The predictive combination of these variables might help clinicians better predict which patients will respond to lithium treatment. Aims To use a combination of transdiagnostic genetic and clinical factors to predict lithium response in patients with bipolar disorder. Method This study utilised genetic and clinical data (n = 1034) collected as part of the International Consortium on Lithium Genetics (ConLi+Gen) project. Polygenic risk scores (PRS) were computed for schizophrenia and major depressive disorder, and then combined with clinical variables using a cross-validated machine-learning regression approach. Unimodal, multimodal and genetically stratified models were trained and validated using ridge, elastic net and random forest regression on 692 patients with bipolar disorder from ten study sites using leave-site-out cross-validation. All models were then tested on an independent test set of 342 patients. The best performing models were then tested in a classification framework. Results The best performing linear model explained 5.1% (P = 0.0001) of variance in lithium response and was composed of clinical variables, PRS variables and interaction terms between them. The best performing non-linear model used only clinical variables and explained 8.1% (P = 0.0001) of variance in lithium response. A priori genomic stratification improved non-linear model performance to 13.7% (P = 0.0001) and improved the binary classification of lithium response. This model stratified patients based on their meta-polygenic loadings for major depressive disorder and schizophrenia and was then trained using clinical data. Conclusions Using PRS to first stratify patients genetically and then train machine-learning models with clinical predictors led to large improvements in lithium response prediction. When used with other PRS and biological markers in the future this approach may help inform which patients are most likely to respond to lithium treatment.

Original languageEnglish (US)
Pages (from-to)219-228
Number of pages10
JournalBritish Journal of Psychiatry
Issue number4
StatePublished - Apr 9 2022


  • Mood stabilisers
  • bipolar affective disorders
  • depressive disorders
  • genetics
  • outcome studies

ASJC Scopus subject areas

  • Psychiatry and Mental health


Dive into the research topics of 'Using polygenic scores and clinical data for bipolar disorder patient stratification and lithium response prediction: machine learning approach'. Together they form a unique fingerprint.

Cite this