Output

SourcePredict result file

File: *.sourcepredict.csv

This csv file contains the predicted proportion of each source in each sample. Like in any classification problem, the predicted source is the greatest proportion.

Example:

+------------------+----------------------+
|                  |      ERR1915662      |
+------------------+----------------------+
| Canis_familiaris |  0.9449678590674971  |
+------------------+----------------------+
|   Homo_sapiens   | 0.027033026106258438 | 
+------------------+----------------------+
|       Soil       | 0.014110223165444446 | 
+------------------+----------------------+
|      unknown     | 0.013888891660799834 |
+------------------+----------------------+

While in this example it is pretty clear that the ERR1915662 sample is likely a dog, you may face situations where it will be less obvious. Looking at the embedding can therefore be useful to decide from which source(s) the sink sample is made up of.

Embedding csv file

This csv file contains the embedding of training in test samples in lower dimensions by TSNE or UMAP

Example:

+-----------------+------------+-------------+--------------+-----------------+
|                 | PC1        | PC2         | labels       | name            |
+-----------------+------------+-------------+--------------+-----------------+
| SRR1175007      | -28.858526 | 0.59231776  | Homo_sapiens | SRR1175007      |
+-----------------+------------+-------------+--------------+-----------------+
| SRR042182       | -22.14415  | -0.47057405 | Homo_sapiens | SRR042182       |
+-----------------+------------+-------------+--------------+-----------------+
| SRR061154       | -30.210106 | -2.0323594  | Homo_sapiens | SRR061154       |
+-----------------+------------+-------------+--------------+-----------------+
| SRR061499       | -25.546652 | 0.27987793  | Homo_sapiens | SRR061499       |
+-----------------+------------+-------------+--------------+-----------------+
| SRR063469       | -22.88011  | 1.1526666   | Homo_sapiens | SRR063469       |
+-----------------+------------+-------------+--------------+-----------------+
| SRR062324       | -25.50832  | -0.25076494 | Homo_sapiens | SRR062324       |
+-----------------+------------+-------------+--------------+-----------------+
| SRR1179037      | -28.779644 | 0.1385772   | Homo_sapiens | SRR1179037      |
+-----------------+------------+-------------+--------------+-----------------+
| SRR061236       | -29.470839 | -0.8973783  | Homo_sapiens | SRR061236       |
+-----------------+------------+-------------+--------------+-----------------+
| SRR061456       | -28.31991  | -0.9834692  | Homo_sapiens | SRR061456       |
+-----------------+------------+-------------+--------------+-----------------+
| SRR1761669      | 4.1411834  | 14.485897   | Homo_sapiens | SRR1761669      |
+-----------------+------------+-------------+--------------+-----------------+
| SRR1761668      | 1.7706155  | 13.6566925  | Homo_sapiens | SRR1761668      |
+-----------------+------------+-------------+--------------+-----------------+
| SRR1761675      | 3.2434833  | 16.020077   | Homo_sapiens | SRR1761675      |
+-----------------+------------+-------------+--------------+-----------------+
| SRR3578625      | 24.127249  | 17.996181   | Soil         | SRR3578625      |
+-----------------+------------+-------------+--------------+-----------------+
| ERR1939165      | 28.738718  | 19.882471   | Soil         | ERR1939165      |
+-----------------+------------+-------------+--------------+-----------------+
| SRR3578645      | 24.138885  | 17.998867   | Soil         | SRR3578645      |
+-----------------+------------+-------------+--------------+-----------------+
| ERR1915662      | -14.770308 | -30.94284   | sink         | ERR1915662      |
+-----------------+------------+-------------+--------------+-----------------+

See the example usage of Sourcepredict for a example of how to plot it.