Predicting positive p53 cancer rescue regions using Most Informative Positive (MIP) active learning.
Many protein engineering problems involve finding mutations that produce proteins with a particular function. Computational active learning is an attractive approach to discover desired biological activities. Traditional active learning techniques have been optimized to iteratively improve classifie...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2009-09-01
|
| Series: | PLoS Computational Biology |
| Online Access: | http://europepmc.org/articles/PMC2742196?pdf=render |
| _version_ | 1819114349390725120 |
|---|---|
| author | Samuel A Danziger Roberta Baronio Lydia Ho Linda Hall Kirsty Salmon G Wesley Hatfield Peter Kaiser Richard H Lathrop |
| author_facet | Samuel A Danziger Roberta Baronio Lydia Ho Linda Hall Kirsty Salmon G Wesley Hatfield Peter Kaiser Richard H Lathrop |
| author_sort | Samuel A Danziger |
| collection | DOAJ |
| description | Many protein engineering problems involve finding mutations that produce proteins with a particular function. Computational active learning is an attractive approach to discover desired biological activities. Traditional active learning techniques have been optimized to iteratively improve classifier accuracy, not to quickly discover biologically significant results. We report here a novel active learning technique, Most Informative Positive (MIP), which is tailored to biological problems because it seeks novel and informative positive results. MIP active learning differs from traditional active learning methods in two ways: (1) it preferentially seeks Positive (functionally active) examples; and (2) it may be effectively extended to select gene regions suitable for high throughput combinatorial mutagenesis. We applied MIP to discover mutations in the tumor suppressor protein p53 that reactivate mutated p53 found in human cancers. This is an important biomedical goal because p53 mutants have been implicated in half of all human cancers, and restoring active p53 in tumors leads to tumor regression. MIP found Positive (cancer rescue) p53 mutants in silico using 33% fewer experiments than traditional non-MIP active learning, with only a minor decrease in classifier accuracy. Applying MIP to in vivo experimentation yielded immediate Positive results. Ten different p53 mutations found in human cancers were paired in silico with all possible single amino acid rescue mutations, from which MIP was used to select a Positive Region predicted to be enriched for p53 cancer rescue mutants. In vivo assays showed that the predicted Positive Region: (1) had significantly more (p<0.01) new strong cancer rescue mutants than control regions (Negative, and non-MIP active learning); (2) had slightly more new strong cancer rescue mutants than an Expert region selected for purely biological considerations; and (3) rescued for the first time the previously unrescuable p53 cancer mutant P152L. |
| first_indexed | 2024-12-22T04:43:53Z |
| format | Article |
| id | doaj.art-c2e4d539b0474c5faa4d94688f6e0e1d |
| institution | Directory Open Access Journal |
| issn | 1553-734X 1553-7358 |
| language | English |
| last_indexed | 2024-12-22T04:43:53Z |
| publishDate | 2009-09-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS Computational Biology |
| spelling | doaj.art-c2e4d539b0474c5faa4d94688f6e0e1d2022-12-21T18:38:39ZengPublic Library of Science (PLoS)PLoS Computational Biology1553-734X1553-73582009-09-0159e100049810.1371/journal.pcbi.1000498Predicting positive p53 cancer rescue regions using Most Informative Positive (MIP) active learning.Samuel A DanzigerRoberta BaronioLydia HoLinda HallKirsty SalmonG Wesley HatfieldPeter KaiserRichard H LathropMany protein engineering problems involve finding mutations that produce proteins with a particular function. Computational active learning is an attractive approach to discover desired biological activities. Traditional active learning techniques have been optimized to iteratively improve classifier accuracy, not to quickly discover biologically significant results. We report here a novel active learning technique, Most Informative Positive (MIP), which is tailored to biological problems because it seeks novel and informative positive results. MIP active learning differs from traditional active learning methods in two ways: (1) it preferentially seeks Positive (functionally active) examples; and (2) it may be effectively extended to select gene regions suitable for high throughput combinatorial mutagenesis. We applied MIP to discover mutations in the tumor suppressor protein p53 that reactivate mutated p53 found in human cancers. This is an important biomedical goal because p53 mutants have been implicated in half of all human cancers, and restoring active p53 in tumors leads to tumor regression. MIP found Positive (cancer rescue) p53 mutants in silico using 33% fewer experiments than traditional non-MIP active learning, with only a minor decrease in classifier accuracy. Applying MIP to in vivo experimentation yielded immediate Positive results. Ten different p53 mutations found in human cancers were paired in silico with all possible single amino acid rescue mutations, from which MIP was used to select a Positive Region predicted to be enriched for p53 cancer rescue mutants. In vivo assays showed that the predicted Positive Region: (1) had significantly more (p<0.01) new strong cancer rescue mutants than control regions (Negative, and non-MIP active learning); (2) had slightly more new strong cancer rescue mutants than an Expert region selected for purely biological considerations; and (3) rescued for the first time the previously unrescuable p53 cancer mutant P152L.http://europepmc.org/articles/PMC2742196?pdf=render |
| spellingShingle | Samuel A Danziger Roberta Baronio Lydia Ho Linda Hall Kirsty Salmon G Wesley Hatfield Peter Kaiser Richard H Lathrop Predicting positive p53 cancer rescue regions using Most Informative Positive (MIP) active learning. PLoS Computational Biology |
| title | Predicting positive p53 cancer rescue regions using Most Informative Positive (MIP) active learning. |
| title_full | Predicting positive p53 cancer rescue regions using Most Informative Positive (MIP) active learning. |
| title_fullStr | Predicting positive p53 cancer rescue regions using Most Informative Positive (MIP) active learning. |
| title_full_unstemmed | Predicting positive p53 cancer rescue regions using Most Informative Positive (MIP) active learning. |
| title_short | Predicting positive p53 cancer rescue regions using Most Informative Positive (MIP) active learning. |
| title_sort | predicting positive p53 cancer rescue regions using most informative positive mip active learning |
| url | http://europepmc.org/articles/PMC2742196?pdf=render |
| work_keys_str_mv | AT samueladanziger predictingpositivep53cancerrescueregionsusingmostinformativepositivemipactivelearning AT robertabaronio predictingpositivep53cancerrescueregionsusingmostinformativepositivemipactivelearning AT lydiaho predictingpositivep53cancerrescueregionsusingmostinformativepositivemipactivelearning AT lindahall predictingpositivep53cancerrescueregionsusingmostinformativepositivemipactivelearning AT kirstysalmon predictingpositivep53cancerrescueregionsusingmostinformativepositivemipactivelearning AT gwesleyhatfield predictingpositivep53cancerrescueregionsusingmostinformativepositivemipactivelearning AT peterkaiser predictingpositivep53cancerrescueregionsusingmostinformativepositivemipactivelearning AT richardhlathrop predictingpositivep53cancerrescueregionsusingmostinformativepositivemipactivelearning |