Help enable the development of safe, effective medicines.
\r\nWhen developing new medicines it is important to identify molecules that are highly active toward their intended targets but not toward other targets that might cause side effects.\r\n The objective of this competition is to identify the best statistical techniques for predicting biological activities of different molecules, both on- and off-target, given numerical descriptors generated from their chemical structures.
\r\nThe challenge is based on 15 molecular activity data sets, each for a biologically relevant target. Each row corresponds to a molecule and contains descriptors derived from that molecule's chemical structure.
\r\nIn addition to the prediction competition, Merck is also hosting a \r\nvisualization challenge with a $2,000 prize for the most insightful and elegant graphical representations of the data.
\r\nPrizes total $40,000.
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "rules", "content": "\r\n\t\r\nYou cannot sign up to Kaggle from multiple accounts and therefore you cannot submit from multiple accounts.
\r\n\r\n Privately sharing code or data outside of teams is not permitted.\r\n It's okay to share code if made available to all participants on the forums.\r\n
\r\n\r\n Team mergers are allowed and can be performed by the team leader. In order to merge, the combined team must have a total submission count less than or equal to the maximum allowed as of the merge date. The maximum allowed is the number of submissions per day multiplied by the number of days the competition has been running.
\r\n
There is no maximum team size.
\r\nYou may submit a maximum of 2 entries per day.
\r\nYou may select up to 5 final submissions for judging.
\r\n\r\n Because the data can be manually decoded and thus an entry can be created by hand, scoring data will be withheld until you have locked down a model.\r\n
\r\n\r\n You must upload all kernels and model files alongside documentation in a single zip file\r\n in order for your entry to be eligible. You can do this at the time of submission, or append\r\n it to your submission at a later time (before the competition end).\r\n
\r\nStart Date: 8/16/2012 9:34 PM UTC
\r\nMerger Deadline: None
\r\nEntry Deadline: None
\r\nEnd Date: 10/16/2012 11:59 PM UTC
\r\n\t\r\nThe Training and Test Sets each consist of 15 biological activity data sets in comma separated value (CSV) format. Each row of data corresponds to a chemical structure represented by molecular descriptors.
\r\nTraining and Test Files
\r\nThe training files are of the form
\r\nThe test files are in the same format with Column 2 removed.
\r\nMolecule IDs and descriptor names are global to all data sets. Thus some molecules will appear in multiple data sets, as will some descriptors.
\r\nThe challenge is to predict the activity value for each molecule/data set combination in the test set. To keep predictions for molecules unique to each data set, a data set identifier has been prepended to each molecule ID (e.g., "ACT1_" or "ACT8_").
\r\nData Set Creation
\r\nFor each activity, the training/test set split is done by dates of testing. That is, the training set consists of compounds assayed by a certain date, and the test set consists of compounds tested after that date. Therefore it is expected that the distribution\r\n of descriptors will not necessarily be the same between the training and test sets.
\r\nAdditional Files
\r\nAlso provided is starter code in R for reading the data sets and producing the naive random forest benchmark, and R code for calculating the R-squared metric. The benchmark result is provided as an example submission file.
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Evaluation", "content": "\r\n
\r\n
\r\n
\r\n
Predictions for activity will be evaluated using the correlation coefficient \\\\(R^2\\\\), averaged over the 15 data sets.
\r\n$$ R^2 = \\frac{1}{15}\\sum_{s=1}^{15} r^2_s $$
\r\nwhere
\r\n$$ r^2_s =\\frac{ [\\sum_{i=1}^{N_s} (x_i-\\bar x)(y_i-\\bar y) ]^2}{ \\sum_{i=1}^{N_s} (x_i-\\bar x)^2 \\sum_{i=1}^{N_s}(y_i-\\bar y)^2 }$$
\r\nwhere \\\\(x\\\\) is the known activity, \\\\(\\bar x\\\\) is the mean of the known activity, \\\\(y\\\\) is the predicted activity, \\\\(\\bar y\\\\) is the mean of the predicted activity, and \\\\(N_s\\\\) is the number of molecules in data set \\\\(s\\\\).
\r\nSample code has been provided to calculate r-squared.
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Visualization Prospect", "content": "Submit interesting and insightful visualizations of the Merck data set for a chance to win the\r\n$2,000 prize! Submissions and voting will open on October 6 and close with the competition on October 14.
\r\nEntries to the Visualization Prospect must be licensed under Creative Commons Attribution-NoDerivs 3.0 (CC BY-ND 3.0). This license allows work to be shared (but not modified)\r\n and used for Commercial and Non-Commercial purposes with attribution to the author.
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Submission Instructions", "content": "Submissions are expected to be in CSV format with 2 columns (MoleculeID, Prediction). Each row will correspond to a particular molecule in a given data set. A complete submission file will have 50,478 rows (including a header row). Files without a header\r\n row will also work.
\r\nPlease download the benchmark (ntree20_basicBenchmark.csv) from the \r\ndata page as an example submission file.
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Winners", "content": "Read the Winner's \r\nAnnouncement on No Free Hunch.
\r\nFirst Place
\r\n![schematic](\"https://storage.googleapis.com/kaggle-media/competitions/kaggle/2975/media/merck_dataset_creation.png\"){kind=link}