Lukas Käll
KTH Royal Institute of Technology

Research interests

Modern biology is to increasing degree dependent on so called high-throughput techniques, i.e. massively parallel experiments that generate a large set of readouts. Examples of such techniques are shotgun proteomics and massively parallel sequencing. A common challenge for these kinds of experiments is that the interpretation of the outcomes, as the individual measurements are of varying quality. We are aiming at increasing the yield and facilitating the interpretation of high-throughput experiments by using different machine learning methods such as support vector machines and dynamical Bayesian networks.

Group members

Heydar Maboudi Afkham, Postdoc
Matthew The, PhD-student

Key publications

• Fast and accurate database searches with MS-GF+ Percolator. V Granholm, S Kim, JCF Navarro, E Sjölund, RD Smith, L Käll. Journal of proteome research (2014), 13 (2), pp 890–897
• Optimized nonlinear gradients for reversed-phase liquid chromatography in shotgun proteomics. L Moruz, P Pichler, T Stranzl, K Mechtler, L Käll. Analytical chemistry (2013) 85 (16), 7777-7785
• Recognizing uncertainty increases robustness and reproducibility of mass spectrometry-based protein inferences O Serang, L Moruz, MR Hoopmann, L Käll. Journal of proteome research (2012), 11 (12), 5586-5591
• On using samples of known protein content to assess the statistical calibration of scores assigned to peptide-spectrum matches in shotgun proteomics. V Granholm, WS Noble, L Käll. Journal of proteome research (2011), 10 (5), 2671-2678
• Training, selection, and robust calibration of retention time models for targeted proteomics. L Moruz, D Tomazela, L Käll. Journal of proteome research( 2010) 9 (10), 5209-5216

External homepage

http://kaell.org/lukas/

Contact

lukas.kall@scilifelab.se

More contact information

https://www.kth.se/profile/lukask/