Lathwal A, et al. - Researchers made a systematic attempt to develop knowledge-based models for prognosticating risk score for colorectal cancer (CRC) patients, which can be used to discriminate high and low-risk patients. On 134 stage III CRC patients, they trained and evaluated the models. The developed models in this work are data-driven where parameters are optimized from the protein concentration obtained from CRC patients. Firstly, using different techniques, they developed multiple linear regression-based models and achieved a maximum HR value of 6.34 with P-value = 0.0032 for a model developed using LassoLars technique. Secondly, a parameter optimization technique was used to develop models and a maximum HR value of 38.13 with P-value 0.0006 was achieved. On predicting favorable/unfavorable survival outcomes, they achieved maximum prediction accuracy value of 71.64%. Addition of clinical factors to this model led to a further enhancement in the performance. When age was added as a variable to the model, the HR improved to 40.11 with P-value as 0.0003 and also boosted the accuracy to 73.13%. Using five-fold cross-validation technique, evaluation of the performance of these models was done. For providing service to the community, a web server ‘CRCRpred’ is developed by the researchers in order to predict risk scores of CRC patients, which is freely available at https://webs.iiitd.edu.in/raghava/crcrpred.