Details on hydrophobicity plots - Revision history

AndrewBangham: /* Fasman et al reviewed different hydrophobicity plots including the data-sieve */

2014-07-30T13:00:13Z

Fasman et al reviewed different hydrophobicity plots including the data-sieve

← Older revision		Revision as of 14:00, 30 July 2014
Line 18:		Line 18:

	These authors compared many different methods and <span style="color:blue;">concluded that the data-sieve was best</span>. However, they found a reason not to recommend it: they didn't understand how it worked.<br><br>		These authors compared many different methods and <span style="color:blue;">concluded that the data-sieve was best</span>. However, they found a reason not to recommend it: they didn't understand how it worked.<br><br>
	Well there you are then. Means verses medians.<br><br>		<span style="color:#A52A2A;">Well there you are then. Means verses medians.</span><br><br>
	'''Actually''', there are some simple computational issues here. At the time computers were only just becoming commonplace and they were not (by current standards) powerful. Computing a running mean is very easy. Choose a window size, ''n'', and start by adding up all the values in the first ''n'' samples to make ''S''. Divide by ''n'' and you have the first point. Now move along by one value - add the new value to ''S'', subtract the first (the outgoing) value from ''S'' and divide by ''n'' again. Add, subtract, divide: easy and the size of ''n'' does not matter. The median is harder, it requires a sort of some type. But I discovered a neat alternative, the recursive median ('m') sieve, indeed I found a family of such filter banks with different kernels 'o', 'c', 'M', 'N', and 'm' and called them all sieves.		'''Actually''', there are some simple computational issues here. At the time computers were only just becoming commonplace and they were not (by current standards) powerful. Computing a running mean is very easy. Choose a window size, ''n'', and start by adding up all the values in the first ''n'' samples to make ''S''. Divide by ''n'' and you have the first point. Now move along by one value - add the new value to ''S'', subtract the first (the outgoing) value from ''S'' and divide by ''n'' again. Add, subtract, divide: easy and the size of ''n'' does not matter. The median is harder, it requires a sort of some type. But I discovered a neat alternative, the recursive median ('m') sieve, indeed I found a family of such filter banks with different kernels 'o', 'c', 'M', 'N', and 'm' and called them all sieves.

AndrewBangham: /* Abstract */

2014-07-30T12:35:43Z

Abstract

← Older revision		Revision as of 13:35, 30 July 2014
Line 9:		Line 9:
	<span style="color:Navy;">''Hydrophobicity plots provide clues to the tertiary structure of proteins (J. Kyte and R. F. Doolittle, 1982, J. Mol. Biol.157, 105; C. Chothia, 1984, Annu. Rev. Biochem.53, 537; T. P. Hopp and K. R. Woods, 1982, Proc. Natl. Acad. Sci. USA78, 3824). To render domains more visible, the raw data are usually smoothed using a running mean of between 5 and 19 amino acids. This type of smoothing still incorporates two disadvantages. First, peculiar residues that do not share the properties of most of the amino acids in the domain may prevent its identification. Second, as a low-pass frequency filter the running mean smoothes sudden transitions from one domain, or phase, to another. Data-sieving is described here as an alternative method for identifying domains within amino acid sequences. The data-sieve is based on a running median '''''and is characterized by a single parameter, the mesh size, which controls its resolution'''''. It is a technique that could be applied to other series data and, '''''in multidimensions, to images''''' in the same way as a median filter.''</span>		<span style="color:Navy;">''Hydrophobicity plots provide clues to the tertiary structure of proteins (J. Kyte and R. F. Doolittle, 1982, J. Mol. Biol.157, 105; C. Chothia, 1984, Annu. Rev. Biochem.53, 537; T. P. Hopp and K. R. Woods, 1982, Proc. Natl. Acad. Sci. USA78, 3824). To render domains more visible, the raw data are usually smoothed using a running mean of between 5 and 19 amino acids. This type of smoothing still incorporates two disadvantages. First, peculiar residues that do not share the properties of most of the amino acids in the domain may prevent its identification. Second, as a low-pass frequency filter the running mean smoothes sudden transitions from one domain, or phase, to another. Data-sieving is described here as an alternative method for identifying domains within amino acid sequences. The data-sieve is based on a running median '''''and is characterized by a single parameter, the mesh size, which controls its resolution'''''. It is a technique that could be applied to other series data and, '''''in multidimensions, to images''''' in the same way as a median filter.''</span>

	~~Actually, for~~ the data-sieve I ~~did use~~ a cascade of medians which was a very inefficient ~~algorithm~~. Subsequently, I switched to a cascade of 1D recursive medians which also ''preserved scale space'' but was ~~hugely~~ ''quicker and is idempotent''.		For the data-sieve I used a cascade of medians which was a very inefficient implementation. Subsequently, I switched to a cascade of 1D recursive medians which also ''preserved scale space'' but was both ''quicker and idempotent''.

			====Fasman et al reviewed different hydrophobicity plots including the data-sieve====

	[[Image:hydrophobicity_plots_Fasman.jpg\|700px]]<br><br>		[[Image:hydrophobicity_plots_Fasman.jpg\|700px]]<br><br>

AndrewBangham: /* Trying to spot regions (lengths) that are up (hydrophobic) and down (hydrophilic) */

2014-07-30T12:27:16Z

Trying to spot regions (lengths) that are up (hydrophobic) and down (hydrophilic)

← Older revision		Revision as of 13:27, 30 July 2014
Line 3:		Line 3:
	Left: Running mean (c.f Gaussian): <span style="color:blue;">no good</span>, Right: Median based Data-sieve: <span style="color:blue;">much better</span>		Left: Running mean (c.f Gaussian): <span style="color:blue;">no good</span>, Right: Median based Data-sieve: <span style="color:blue;">much better</span>
	[[Image:HydrophobicityPlot_Bacteriorhodopsin.png\|700px]]		[[Image:HydrophobicityPlot_Bacteriorhodopsin.png\|700px]]
	Bangham, J.A. (1988) Data-sieving hydrophobicity plots. Anal. Biochem. 174, 142–145<br>		<span style="color:Navy;">Bangham, J.A. (1988) Data-sieving hydrophobicity plots. Anal. Biochem. 174, 142–145</span><br>

	====Abstract====		====Abstract====

AndrewBangham: /* Abstract */

2014-07-30T12:26:01Z

Abstract

← Older revision		Revision as of 13:26, 30 July 2014
Line 5:		Line 5:
	Bangham, J.A. (1988) Data-sieving hydrophobicity plots. Anal. Biochem. 174, 142–145<br>		Bangham, J.A. (1988) Data-sieving hydrophobicity plots. Anal. Biochem. 174, 142–145<br>

	==Abstract==		====Abstract====

	''Hydrophobicity plots provide clues to the tertiary structure of proteins (J. Kyte and R. F. Doolittle, 1982, J. Mol. Biol.157, 105; C. Chothia, 1984, Annu. Rev. Biochem.53, 537; T. P. Hopp and K. R. Woods, 1982, Proc. Natl. Acad. Sci. USA78, 3824). To render domains more visible, the raw data are usually smoothed using a running mean of between 5 and 19 amino acids. This type of smoothing still incorporates two disadvantages. First, peculiar residues that do not share the properties of most of the amino acids in the domain may prevent its identification. Second, as a low-pass frequency filter the running mean smoothes sudden transitions from one domain, or phase, to another. Data-sieving is described here as an alternative method for identifying domains within amino acid sequences. The data-sieve is based on a running median '''''and is characterized by a single parameter, the mesh size, which controls its resolution'''''. It is a technique that could be applied to other series data and, '''''in multidimensions, to images''''' in the same way as a median filter.''		<span style="color:Navy;">''Hydrophobicity plots provide clues to the tertiary structure of proteins (J. Kyte and R. F. Doolittle, 1982, J. Mol. Biol.157, 105; C. Chothia, 1984, Annu. Rev. Biochem.53, 537; T. P. Hopp and K. R. Woods, 1982, Proc. Natl. Acad. Sci. USA78, 3824). To render domains more visible, the raw data are usually smoothed using a running mean of between 5 and 19 amino acids. This type of smoothing still incorporates two disadvantages. First, peculiar residues that do not share the properties of most of the amino acids in the domain may prevent its identification. Second, as a low-pass frequency filter the running mean smoothes sudden transitions from one domain, or phase, to another. Data-sieving is described here as an alternative method for identifying domains within amino acid sequences. The data-sieve is based on a running median '''''and is characterized by a single parameter, the mesh size, which controls its resolution'''''. It is a technique that could be applied to other series data and, '''''in multidimensions, to images''''' in the same way as a median filter.''</span>

	Actually, for the data-sieve I did use a cascade of medians which was a very inefficient algorithm. Subsequently, I switched to a cascade of 1D recursive medians which also ''preserved scale space'' but was hugely ''quicker and is idempotent''.		Actually, for the data-sieve I did use a cascade of medians which was a very inefficient algorithm. Subsequently, I switched to a cascade of 1D recursive medians which also ''preserved scale space'' but was hugely ''quicker and is idempotent''.

AndrewBangham: /* Trying to spot regions (lengths) that are up (hydrophobic) and down (hydrophilic) */

2014-07-30T12:24:21Z

Trying to spot regions (lengths) that are up (hydrophobic) and down (hydrophilic)

← Older revision		Revision as of 13:24, 30 July 2014
Line 3:		Line 3:
	Left: Running mean (c.f Gaussian): <span style="color:blue;">no good</span>, Right: Median based Data-sieve: <span style="color:blue;">much better</span>		Left: Running mean (c.f Gaussian): <span style="color:blue;">no good</span>, Right: Median based Data-sieve: <span style="color:blue;">much better</span>
	[[Image:HydrophobicityPlot_Bacteriorhodopsin.png\|700px]]		[[Image:HydrophobicityPlot_Bacteriorhodopsin.png\|700px]]
	Bangham, J.A. (1988) Data-sieving hydrophobicity plots. Anal. Biochem. 174, 142–145<br>~~<br>~~		Bangham, J.A. (1988) Data-sieving hydrophobicity plots. Anal. Biochem. 174, 142–145<br>


	==Abstract==		==Abstract==

AndrewBangham: /* Don't give up, get motivated */

2014-06-18T18:06:42Z

Don't give up, get motivated

← Older revision		Revision as of 19:06, 18 June 2014
Line 24:		Line 24:
	==Don't give up, get motivated==		==Don't give up, get motivated==
	The good thing was that I was motivated to prove exactly how it worked and why it was so good - I moved to Computing Sciences and published everything in computing journals. Where computer scientists would find it - or not - think MSER literature.<br><br>		The good thing was that I was motivated to prove exactly how it worked and why it was so good - I moved to Computing Sciences and published everything in computing journals. Where computer scientists would find it - or not - think MSER literature.<br><br>
	I was particularly excited about the possibility that the new set of scale-space preserving transforms might shine a light on brain vision. I attended conferences on vision - the amazing Horace Barlow's retirement bash, the 6 month extravaganza: ''Computer Vision'' at the Isaac Newton Institute for Mathematical Sciences (University of Cambridge, United Kingdom) July to December, 1993. Organised by scientists that I greatly admire: David Mumford (Harvard), Andrew Blake, Brian Ripley (Oxford) etc. However, it was difficult. A typical remark was 'But mathematicians have shown that Gaussian Filter banks are unique ...' The focus then was on anisotropic diffusion.<br><br>		I was particularly excited about the possibility that the new set of scale-space preserving transforms might shine a new and different light on brain vision. Fourier transforms, wavelets, Gabor filters etc. may not always be the right light (OK I agree, the 'standard' explanations are very consistent with the evidence, but sometimes one should test and reject alternatives). I attended conferences on vision - the amazing Horace Barlow's retirement bash, the 6 month extravaganza: ''Computer Vision'' at the Isaac Newton Institute for Mathematical Sciences (University of Cambridge, United Kingdom) July to December, 1993. Organised by scientists that I greatly admire: David Mumford (Harvard), Andrew Blake, Brian Ripley (Oxford) etc. However, it was difficult. A typical remark was 'But mathematicians have shown that Gaussian Filter banks are unique ...' The focus then was on anisotropic diffusion.<br><br>
	I was <span style="color:blue;">proving</span> that diffusion filters are not unique. <br><br>		I was <span style="color:blue;">proving</span> that diffusion filters are not unique. <br><br>
	The non-linear filters I was describing held the promise of extremely useful properties in computer vision - as it turns out one example would be the MSER's. Some of their success will be down to the front end processing preserving scale space.		The non-linear filters I was describing held the promise of extremely useful properties in computer vision - as it turns out one example would be the MSER's. Some of their success will be down to the front end processing preserving scale space.

AndrewBangham: /* Abstract */

2014-06-18T17:52:49Z

Abstract

← Older revision		Revision as of 18:52, 18 June 2014
Line 16:		Line 16:
	Left: Running mean (c.f Gaussian): <span style="color:blue;">no good</span>, Right: Median based Data-sieve: <span style="color:blue;">much better</span><br><br>		Left: Running mean (c.f Gaussian): <span style="color:blue;">no good</span>, Right: Median based Data-sieve: <span style="color:blue;">much better</span><br><br>

	These authors compared many different methods and <span style="color:blue;">concluded that the data-sieve was best</span>. However, they found a reason not to recommend it: ~~that~~ they didn't understand how it worked.<br><br>		These authors compared many different methods and <span style="color:blue;">concluded that the data-sieve was best</span>. However, they found a reason not to recommend it: they didn't understand how it worked.<br><br>
	Well there you are then. Means verses medians.<br><br>		Well there you are then. Means verses medians.<br><br>
	'''Actually''', there are some simple computational issues here. At the time computers were only just becoming commonplace and they were not (by current standards) powerful. Computing a running mean is very easy. Choose a window size, ''n'', and start by adding up all the values in the first ''n'' samples to make ''S''. Divide by ''n'' and you have the first point. Now move along by one value - add the new value to ''S'', subtract the first (the outgoing) value from ''S'' and divide by ''n'' again. Add, subtract, divide: easy and the size of ''n'' does not matter. The median is harder, it requires a sort of some type. But I discovered a neat alternative, the recursive median ('m') sieve, indeed I found a family of such filter banks with different kernels 'o', 'c', 'M', 'N', and 'm' and called them all sieves.		'''Actually''', there are some simple computational issues here. At the time computers were only just becoming commonplace and they were not (by current standards) powerful. Computing a running mean is very easy. Choose a window size, ''n'', and start by adding up all the values in the first ''n'' samples to make ''S''. Divide by ''n'' and you have the first point. Now move along by one value - add the new value to ''S'', subtract the first (the outgoing) value from ''S'' and divide by ''n'' again. Add, subtract, divide: easy and the size of ''n'' does not matter. The median is harder, it requires a sort of some type. But I discovered a neat alternative, the recursive median ('m') sieve, indeed I found a family of such filter banks with different kernels 'o', 'c', 'M', 'N', and 'm' and called them all sieves.

AndrewBangham: /* Abstract */

2014-06-18T17:51:49Z

Abstract

← Older revision		Revision as of 18:51, 18 June 2014
Line 10:		Line 10:
	''Hydrophobicity plots provide clues to the tertiary structure of proteins (J. Kyte and R. F. Doolittle, 1982, J. Mol. Biol.157, 105; C. Chothia, 1984, Annu. Rev. Biochem.53, 537; T. P. Hopp and K. R. Woods, 1982, Proc. Natl. Acad. Sci. USA78, 3824). To render domains more visible, the raw data are usually smoothed using a running mean of between 5 and 19 amino acids. This type of smoothing still incorporates two disadvantages. First, peculiar residues that do not share the properties of most of the amino acids in the domain may prevent its identification. Second, as a low-pass frequency filter the running mean smoothes sudden transitions from one domain, or phase, to another. Data-sieving is described here as an alternative method for identifying domains within amino acid sequences. The data-sieve is based on a running median '''''and is characterized by a single parameter, the mesh size, which controls its resolution'''''. It is a technique that could be applied to other series data and, '''''in multidimensions, to images''''' in the same way as a median filter.''		''Hydrophobicity plots provide clues to the tertiary structure of proteins (J. Kyte and R. F. Doolittle, 1982, J. Mol. Biol.157, 105; C. Chothia, 1984, Annu. Rev. Biochem.53, 537; T. P. Hopp and K. R. Woods, 1982, Proc. Natl. Acad. Sci. USA78, 3824). To render domains more visible, the raw data are usually smoothed using a running mean of between 5 and 19 amino acids. This type of smoothing still incorporates two disadvantages. First, peculiar residues that do not share the properties of most of the amino acids in the domain may prevent its identification. Second, as a low-pass frequency filter the running mean smoothes sudden transitions from one domain, or phase, to another. Data-sieving is described here as an alternative method for identifying domains within amino acid sequences. The data-sieve is based on a running median '''''and is characterized by a single parameter, the mesh size, which controls its resolution'''''. It is a technique that could be applied to other series data and, '''''in multidimensions, to images''''' in the same way as a median filter.''

	Actually, for the data-sieve I did use a cascade of medians which was a very inefficient algorithm. Subsequently, I switched to a cascade of 1D recursive medians which also preserved scale space but was hugely quicker and is idempotent.		Actually, for the data-sieve I did use a cascade of medians which was a very inefficient algorithm. Subsequently, I switched to a cascade of 1D recursive medians which also ''preserved scale space'' but was hugely ''quicker and is idempotent''.

	[[Image:hydrophobicity_plots_Fasman.jpg\|700px]]<br><br>		[[Image:hydrophobicity_plots_Fasman.jpg\|700px]]<br><br>

AndrewBangham: /* Don't give up, get motivated */

2014-06-18T17:50:21Z

Don't give up, get motivated

← Older revision		Revision as of 18:50, 18 June 2014
Line 26:		Line 26:
	I was particularly excited about the possibility that the new set of scale-space preserving transforms might shine a light on brain vision. I attended conferences on vision - the amazing Horace Barlow's retirement bash, the 6 month extravaganza: ''Computer Vision'' at the Isaac Newton Institute for Mathematical Sciences (University of Cambridge, United Kingdom) July to December, 1993. Organised by scientists that I greatly admire: David Mumford (Harvard), Andrew Blake, Brian Ripley (Oxford) etc. However, it was difficult. A typical remark was 'But mathematicians have shown that Gaussian Filter banks are unique ...' The focus then was on anisotropic diffusion.<br><br>		I was particularly excited about the possibility that the new set of scale-space preserving transforms might shine a light on brain vision. I attended conferences on vision - the amazing Horace Barlow's retirement bash, the 6 month extravaganza: ''Computer Vision'' at the Isaac Newton Institute for Mathematical Sciences (University of Cambridge, United Kingdom) July to December, 1993. Organised by scientists that I greatly admire: David Mumford (Harvard), Andrew Blake, Brian Ripley (Oxford) etc. However, it was difficult. A typical remark was 'But mathematicians have shown that Gaussian Filter banks are unique ...' The focus then was on anisotropic diffusion.<br><br>
	I was <span style="color:blue;">proving</span> that diffusion filters are not unique. <br><br>		I was <span style="color:blue;">proving</span> that diffusion filters are not unique. <br><br>
	The non-linear filters I was describing held the promise of extremely useful properties in computer vision - as it turns out one example ~~could~~ be the MSER's ~~where~~ the front end processing ~~preserves~~ scale space.		The non-linear filters I was describing held the promise of extremely useful properties in computer vision - as it turns out one example would be the MSER's. Some of their success will be down to the front end processing preserving scale space.

AndrewBangham: /* Don't give up, get motivated */

2014-06-18T17:49:07Z

Don't give up, get motivated

← Older revision		Revision as of 18:49, 18 June 2014
Line 24:		Line 24:
	==Don't give up, get motivated==		==Don't give up, get motivated==
	The good thing was that I was motivated to prove exactly how it worked and why it was so good - I moved to Computing Sciences and published everything in computing journals. Where computer scientists would find it - or not - think MSER literature.<br><br>		The good thing was that I was motivated to prove exactly how it worked and why it was so good - I moved to Computing Sciences and published everything in computing journals. Where computer scientists would find it - or not - think MSER literature.<br><br>
	I was particularly excited about the possibility that the new set of scale-space preserving transforms might shine a light on brain vision. I attended conferences on vision - the amazing Horace Barlow's retirement bash, the 6 month extravaganza: ''Computer Vision'' at the Isaac Newton Institute for Mathematical Sciences (University of Cambridge, United Kingdom) July to December, 1993. Organised by scientists that I greatly admire: David Mumford (Harvard), Andrew Blake, Brian Ripley (Oxford)etc. However, it was difficult. A typical remark was 'But mathematicians have shown that Gaussian Filter banks are unique ...' The focus was anisotropic diffusion.<br><br>		I was particularly excited about the possibility that the new set of scale-space preserving transforms might shine a light on brain vision. I attended conferences on vision - the amazing Horace Barlow's retirement bash, the 6 month extravaganza: ''Computer Vision'' at the Isaac Newton Institute for Mathematical Sciences (University of Cambridge, United Kingdom) July to December, 1993. Organised by scientists that I greatly admire: David Mumford (Harvard), Andrew Blake, Brian Ripley (Oxford) etc. However, it was difficult. A typical remark was 'But mathematicians have shown that Gaussian Filter banks are unique ...' The focus then was on anisotropic diffusion.<br><br>
	I was <span style="color:blue;">proving</span> that diffusion filters are not unique. <br><br>		I was <span style="color:blue;">proving</span> that diffusion filters are not unique. <br><br>
	The non-linear filters I was describing held the promise of extremely useful properties in computer vision - as it turns out one example could be the MSER's where the front end processing preserves scale space.		The non-linear filters I was describing held the promise of extremely useful properties in computer vision - as it turns out one example could be the MSER's where the front end processing preserves scale space.