a protein sequences example added

# Usage instructions

## Help

Enter in terminal

    bin/pcfg_scan --help

to print full list of program options.

## Scanning example

Directory `./example` contains files:

- `hets.lex.wcfg` (lexical part of grammar)

- `hets.struct.wcfg` (structural part of grammar)

- `hets.fasta` (example sequences to be parsed)

- `hets.cmap` (list of contact pairs)

- `hets-scan.conf` (configuration file used in this example)

- `hets-evolve.conf` (configuration file used in the next example)

- `hets-galib.conf` (configuration file also used in the next example)

Let's run the scanner on these files:

    bin/pcfg_scan --conf ./example/hets-scan.conf --out trees.txt --tree

Here is the meaning of options used above:

- `--conf ...` specifies configuration file

- `--out ...` specifies name of output file

- `--tree` orders scanner to print parse trees

There are also other options, which are loaded from configuration file.

All options except `--conf` can be set in either way.

- `clist=..` specifies file containing contact pairs

- `fasta=...` specifies file containing input protein sequences

- `lex=...` specifies file containing lexical rules of grammar

- `struct=...` specifies file containing structural part of grammar

- `null=...` chooses null model

- `winmax=...` sets maximum window size

- `winmin=...` sets minimum window size

- `outdir=...` specifies directory where output file(s) are created

Scanner should then produce output file `trees.txt` in directory `./example`.

## Learning example

Please read the paragraph about scanning first.

In this example we use the same files as in previous example,

except the configuration file used now is `./example/hets-evolve.conf`.

Let's run the learning algorithm on these files:

    bin/pcfg_evolve --conf ./example/hets-evolve.conf --seed 1571856417

Option `--seed ...` explicitly specifies seed for pseudorandom numbers used,

which is useful for reproducing results.

There are some options which are loaded from the configuration file,

but were not present in the previous example.

- `obj=G_MX` specifies objective function, here we specified that we are looking for the most likely grammar(G) given contacts (M) and sequences (X)

- `galib-conf=...` specifies configuration file with options specific to GALIB library

- `sharing-cutoff=...` specifies sharing cutoff

- `grammar-flush-frequency=...` specifies how often program saves the current best grammar to file

Scanner should then produce output files in directory `./example`:

- `1571856417-galib.csv`

- `1571856417-50-0-lex.wcfg`

- `1571856417-50-0-struct.wcfg`

- `1571856417-final-0-lex.wcfg`

- `1571856417-final-0-struct.wcfg`

- `1571856417-galib.conf`

- `1571856417-pcfg_evolve.conf`

## Common options

### `--clist arg`

Switch to Viterbi parsing.

By default, probabilities are added, which corresponds to Baum-Welch mode of operation, wihch gives overall probability of the sequence in grammar. When this option is enabled, during parsing probabilities are combined with `max` function, giving probability of the most likely derivation. Notably, option `--tree` in `pcfg_scan` operates as if this option was set, among other things.

By default, probabilities are added, which corresponds to Baum-Welch mode of operation, wihch gives overall probability of the sequence in grammar. When this option is enabled, during parsing probabilities are combined with `max` function, giving probability of the most likely derivation. Notably, option `--tree` in `pcfg_scan` always operates as if this option was set.

## Options specific for `pcfg_scan`

clist=./example/hets.cmap

fasta=./example/hets.fasta

lex=./example/hets.lex.wcfg

struct=./example/hets.struct.wcfg

outdir=./example

obj=G_MX

galib-conf=./example/hets-galib.conf

sharing-cutoff=20.0	# 1.0 - default

grammar-flush-frequency=50	# 0 (only after finish) - default

# sample settings for GAlib applications

# GAlib expects parameters in name-value pairs.  The name should be a single

# string (no whitespace allowed).  The value should be of a type appropriate

# for the named parameter.  Anything after a # character will be ignored.

# The file must end with a blank line.  If you specify parameters that depend

# on other parameters, the last parameter will override the first or act on

# data modified by the first (the parameters are applied in the order they 

# are listed).

minimaxi                    -1

number_of_generations       100

convergence_percentage      1.001

generations_to_convergence  10

crossover_probability       0.90

mutation_probability        0.01

population_size             100

replacement_percentage      0.50

number_of_best              1

score_frequency             1

flush_frequency             100

select_scores               31

record_diversity            1

clist=./example/hets.cmap

fasta=./example/hets.fasta

lex=./example/hets.lex.wcfg

struct=./example/hets.struct.wcfg

outdir=./example

null=p0

winmax=21

winmin=21

>jr1_Melbi2_547107_e_gw1.

8	#	16

10	#	14

17	#	21

>jr1_Aurpu_var_mel1_81883

8	#	16

10	#	14

17	#	21

>jr1_Aaoar1_429219_fgenes

8	#	16

10	#	14

17	#	21

>jr1_Antav1_424285_fgenes

8	#	16

10	#	14

17	#	21

>r2_682456863_gb_KFZ14600

8	#	16

10	#	14

17	#	21

>jr2_Amore1_33010_fgenesh

8	#	16

10	#	14

17	#	21

>jr2_Aspsy1_33517_fgenesh

8	#	16

10	#	14

17	#	21

>jr1_Thihy1_493338_fgenes

8	#	16

10	#	14

17	#	21

>jr2_Aciri1_iso_159232_fg

8	#	16

10	#	14

17	#	21

>jr2_Clagr3_6461_CLAGR_00

8	#	16

10	#	14

17	#	21

>jr1_Thiar1_783280_fgenes

8	#	16

10	#	14

17	#	21

>r2_596672318_ref_XP_0072

8	#	16

10	#	14

17	#	21

>jr2_Cadsp1_662565_estExt

8	#	16

10	#	14

17	#	21

>jr2_Coclu2_34207_gm1.295

8	#	16

10	#	14

17	#	21

>jr2_Cocsa1_192976_estExt

8	#	16

10	#	14

17	#	21

>jr2_Gloci1_1938808_MIX33

8	#	16

10	#	14

17	#	21

>jr1_Necha2_48298_e_gw1.2

8	#	16

10	#	14

17	#	21

>jr1_TriviGv29_8_2_28501_

8	#	16

10	#	14

17	#	21

>r1_671156951_ref_XP_0087

8	#	16

10	#	14

17	#	21

>jr2_Clafu1_191299_scf718

8	#	16

10	#	14

17	#	21