Main Menu
About PconsC
Database status
Resource status

Random forest prediction of contacts in protein tertiary structure from primary structure.

PconsC takes protein sequence as input data. Jackhmmer and HHblits are then used to generate alignments at various levels of confidence of homology. For each alignment, contact predictions are made using PSICOV and plmDCA. The combined output is then used as basis for a random forest evaluation, which produces the final result.

More information available on:

PconsC is also available from github:


Phone: +46852481531
Email: click here to contact Arne
Address: Arne Elofsson, Science for Life Laboratory, Box 1031, 17121 Solna, Sweden

For the website: please contact nanjiang.shu at

Skwark, M. J., Abdel-Rehim, A., & Elofsson, A. (2013). "PconsC: combination of direct information methods and alignments improves contact prediction". Bioinformatics, 29(14), 1815-1816.0

Skwark M.J., Raimondi D., Michel M.. and Elofsson A. "Improved contact predictions using the recognition of protein like contact patterns".
© Arne Elofsson

SciLifeLab Logotype BILS Logotype
SeRC Logotype Stockholm University logotype